Sometimes it takes just one human being to tip the

scales and change the course of history. At some point UN-HABITAT

in the year 2007, that human being will either move

to a city or be born in one. The event itself will go
unnoticed but demographers watching urban trends STATE OF THE WORLD’S

cities
will mark it as the moment when the world entered a
new urban millennium, a period in which, for the
first time in history, the majority of the world’s people
will live in cities. The year 2007 will also see the
number of slum dwellers in the world cross the one
billion mark – when one in every three city residents
will live in a slum.

Although poverty remains a primarily rural

phenomenon, large sections of the urban population
in developing countries are suffering from extreme
levels of deprivation that are often even more
2006/7
debilitating than those experienced by the rural poor.
The Millennium
UN-HABITAT analyses reflected in this Report
Development Goals and

S TAT E O F T H E WO R L D ’ S C I T I E S 2 0 0 6 / 7
show that the incidence of disease and mortality is
much higher in slums than in non-slum urban areas, Urban Sustainability:
and in some cases, such as HIV prevalence and other 30 Years of Shaping
health indicators, is equal to or even higher than in the Habitat Agenda
rural areas. These disparities are often not reflected in
national statistics, which mask the deprivation
experienced in poor urban neighbourhoods. The
findings in this Report reveal “a tale of two cities
within one city”, where non-slum populations enjoy
good health and education, while slum communities
suffer from both poor health and lack of
opportunities.

This edition of the State of the World’s Cities Report

provides an overview of a range of issues that link
cities, slums and the Millennium Development
Goals. It makes clear that the global fight against
poverty – encapsulated in the Millennium
Development Goals – is heavily dependent on how
well cities perform.

The Report highlights three inter-related issues:

• The Millennium Development Goals provide an
apt framework for linking the opportunities
provided by cities with improved quality of life;
• The achievement of the Goals heavily depends on
governments’ capacity to speed up progress and
reverse current trends on slum formation;
• The achievement of the Goals depends on
governments' capacity to speed up progress in
reducing urban poverty and inequality and in
reversing current trends in slum formation.

www.unhabitat.org

UN-HABITAT
First published by Earthscan in the UK and USA in 2006 for and on behalf of the United Nations Human Settlements Programme (UN-HABITAT)

Copyright © United Nations Human Settlements Programme, 2006

All rights reserved

United Nations Human Settlements Programme (UN-HABITAT)

P.O. Box 30030, Nairobi, Kenya
Tel: +254 20 7621 234
Fax: +254 20 7624 266/7
Website: www.unhabitat.org

DISCLAIMER

The designations employed and the presentation of the material in this report do not imply the expression of any opinion whatsoever on the part of the
Secretariat of the United Nations concerning the legal status of any country, territory, city or area, or of its authorities, or concerning delimitation of its
frontiers or boundaries, or regarding its economic system or degree of development. The analysis, conclusions and recommendations of this report do
not necessarily reflect the views of the United Nations Human Settlements Programme or its Governing Council.

HS/814/06E (paperback)
HS/815/06E (hardback)

ISBN-10: 1-84407-378-5 (Earthscan paperback) 92-1-131811-4 (UN-HABITAT paperback)

ISBN-13: 978-1-84407-378-8 (Earthscan paperback) 978-92-1-131811-1 (UN-HABITAT paperback)
ISBN-10: 1-84407-379-3 (Earthscan hardback) 92-1-131812-2 (UN-HABITAT hardback)
ISBN-13: 978-1-84407-379-5 (Earthscan hardback) 978-92-1-131812-8 (UN-HABITAT hardback)

Design and layout by Michael Jones Software, Nairobi, Kenya

Printed and bound in Malta by Gutenberg Press Ltd

For a full list of Earthscan publications contact:

Earthscan
8–12 Camden High Street
London, NW1 0JH, UK
Tel: +44 (0)20 7387 8558
Fax: +44 (0)20 7387 8998
E-mail: earthinfo@earthscan.co.uk
Website: www.earthscan.co.uk

22883 Quicksilver Drive, Sterling, VA 20166-2012, USA

Earthscan is an imprint of James & James (Science Publishers) Ltd and publishes in association with the International Institute for Environment and
Development

A catalogue record for this book is available from the British Library

Library of Congress Cataloging-in-Publication Data

[to follow]
THE STATE OF THE WORLD’S CITIES REPORT 2006/2007
30 Years of Shaping the Habitat Agenda

United Nations Human Settlements Programme

S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Forword Introduction
The 2006/2007 edition of the State of the World’s Cities marks
two important milestones: the dawn of the urban millennium
in 2007 and the 30th anniversary of the first Habitat
Conference held in Vancouver in June 1976, which placed
“urbanisation” on the global development agenda. This publi-
cation also marks a less triumphal moment in history. Thirty
years after the world’s governments first pledged to do more for
cities, almost one-third of the world’s urban population lives in
slums, most of them without access to decent housing or basic
services and where disease, illiteracy and crime are rampant.
Since its establishment in 1979, the United Nations Human
Settlements Programme (UN-HABITAT) has continued to
highlight the important role and contribution of cities in fos-
tering economic and human development. Understanding the
complex social, cultural and economic dynamics of cities and
urbanisation is more important now than ever before as we
strive to attain internationally agreed development goals. In a
rapidly urbanizing world attaining these goals will require poli-
cies and strategies based on clear and accurate data on the
human settlements conditions and trends in each country.
This edition of the State of the World’s Cities advances this
objective by breaking new ground in the area of urban data col-
lection, analysis and dissemination. For the first time in the his-
tory of the United Nations, urban data is reported here at slum
and non-slum levels, going far beyond the traditional urban-
rural dichotomy. UN-HABITAT’s intra-urban data analysis –
involving disaggregated data for more than 200 cities around
the world – takes this work further and provides detailed evi-
dence of urban inequalities in the areas of health, education,
employment and other key indicators. The implications are sig-
nificant for the attainment of Millennium Development Goals
as we can no longer assume that the urban poor are better off
than their rural counterparts, or that all urban dwellers are able
to benefit from basic services by virtue of proximity.
UN-HABITAT has led the drive for urban indicators since
1991 by working with other United Nations agencies and
external partners to consistently refine methods for data collec-
tion and analysis and to better inform our common quest for
“adequate shelter for all” and “sustainable human settlements
development in an urbanizing world” - the twin goals of the
Habitat Agenda adopted by the world’s governments in
ii
Istanbul in 1996. With the adoption of the Millennium
Declaration by the world’s leaders in 2000, much of this work
is now focused on monitoring progress in attaining
Millennium Development Goal 7, target 11 on improving the
lives of at least 100 million slum dwellers by 2020. This task
requires the analysis of how well cities are doing and of the
actual living conditions of the urban poor. Data for this report
comes primarily from Phase III of UN-HABITAT’s Urban
Indicators Programme that compiles global, regional, country
and household-level data of specific relevance to the Habitat
Agenda and the Millennium Development Goals.
This Report clearly shows how shelter conditions have a direct
impact on human development, including child mortality, edu-
cation and employment. The correlation between a poor living
environment, characterized by one or more shelter depriva-
tions, and poor performance on key indicators of the
Millennium Development Goals underscores the assertion that
“where we live matters”. The findings of this report are unfold-
ing a new urban reality that needs to be urgently addressed by
pro-poor and gender-sensitive urban policies and legislation.
Finally, as the international community celebrates Vancouver +
30, it should also reflect on the important lessons learned in
urban development and the need to reduce inequalities within
cities. Cities present an unparalleled opportunity for the simul-
taneous attainment of most if not all of the internationally
agreed development goals. Interventions in, for example, pro-
poor water and sanitation, have immediate positive knock-on
effects in terms of improved health, nutrition, disease preven-
tion and the environment. However, unless such concerted
action is taken to redress urban inequalities cities may well
become the predominant sites of deprivation, social exclusion
and instability worldwide.
Anna K. Tibaijuka
Under-Secretary-General and Executive Director,
UN-HABITAT
iii
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Overview
S
ometimes it takes just one human being to tip the
scales and change the course of history. At some point
in the year 2007, that human being will either move
to a city or be born in one. The event itself will go
unnoticed but demographers watching urban trends
will mark it as the moment when the world entered a new
urban millennium, a period in which, for the first time in his-
tory, the majority of the world’s people will live in cities.
The year 2007 will also see the number of slum dwellers in
the world cross the one billion mark – when one in every three
city residents will live in inadequate housing with no or few
basic services. This statistic may be reported in newspaper
headlines, but it is still not yet clear how it will influence gov-
ernment policies and actions, particularly in relation to
Millennium Development Goal 7, target 11: by 2020, to have
improved the lives of at least 100 million slum dwellers.
■ Cities, Slums and the Millennium
Development Goals
Three important trends characterize the urbanization process
in this new urban era. Firstly, the biggest cities in the world will
be found mainly in the developing world. “Metacities” – mas-
sive conurbations of more than 20 million people, above and
beyond the scale of megacities – are now gaining ground in
Asia, Latin America and Africa. These cities are home to only 4
per cent of the world’s population and most have grown at the
relatively slow rate of about 1.5 per cent annually. However, the
sheer size of these urban agglomerations points to the growth
of city-regions and “metropolitanization” that call for more
polycentric forms of urban governance and management and
stronger inter-municipal relations. The scale of environmental
impact of metacities and megacities on their hinterlands is also
significant and is likely to be a cause for concern in coming
decades.
Secondly, despite the emergence of metacities, the majority
of urban migrants will be moving to small towns and cities of
less than one million inhabitants. Already, more than half of
the world’s urban population lives in cities of fewer than
500,000 inhabitants, and almost one-fifth lives in cities of
between 1 and 5 million inhabitants. These intermediate cities
are predicted to grow at a faster rate than any other type of city.
Natural population increase, rather than rural-to-urban migra-
tion, is becoming a more significant contributor to urban
iv
growth in many regions, as is reclassification of rural areas into
urban areas. However, the relative absence of infrastructure,
such as roads, water supply and communication facilities, in
many small and intermediate-sized cities makes these cities less
competitive locally, nationally and regionally, and leads to a
lower quality of life for their citizens.
Thirdly, cities of the developing world will absorb 95 per
cent of urban growth in the next two decades, and by 2030,
will be home to almost 4 billion people, or 80 per cent of the
world’s urban population. After 2015, the world’s rural popu-
lation will begin to shrink as urban growth becomes more
intense in cities of Asia and Africa, two regions that are set to
host the world’s largest urban populations in 2030, 2.66 billion
and 748 million, respectively. Urban poverty and inequality
will characterize many cities in the developing world, and
urban growth will become virtually synonymous with slum for-
mation in some regions. Asia is already home to more than half
of the world’s slum population (581 million) – followed by
sub-Saharan Africa (199 million) and Latin America and the
Caribbean (134 million). Sub-Saharan Africa has both the
highest annual urban growth rate and the highest slum growth
rate in the world, 4.58 per cent and 4.53 per cent, respectively,
more than twice the world average. The continued threat of
conflict in several African countries is a significant contributing
factor in the proliferation of slums in the region’s urban areas.
The prolonged crisis in Southern Sudan, for instance, has led
to the mass exodus of rural communities to the capital
Khartoum, which accommodated almost half of the more than
6 million internally displaced persons in the country in the late
1990s. These trends will most likely concern policymakers in
the developing world as they confront the reality of growing
inequality and poverty in their cities.
The good news is that urbanization can also be a positive
force for human development; countries that are highly urban-
ized tend to have higher incomes, more stable economies,
stronger institutions and are better able to withstand the
volatility of the global economy. In both developed and devel-
oping countries, cities generate a disproportionate share of
gross domestic product and provide extensive opportunities for
employment and investment. However, evidence suggests that
despite the enormous potential of cities to bring about prosper-
ity, the wealth generated by cities does not automatically lead
to poverty reduction; on the contrary, in many cities, inequali-
ties between the rich and the poor have grown, as have the sizes
and proportions of slum populations.
Although poverty remains a primarily rural phenomenon,
urban poverty is becoming a severe, pervasive - and largely unac-
knowledged - feature of urban life. Large sections of the popula-
tion in urban areas are suffering from extreme levels of depriva-
tion that are often even more debilitating than those experienced
by the rural poor. UN-HABITAT analyses reflected in this Report
show that the incidence of disease and mortality is much higher
in slums than in non-slum urban areas, and in some cases, such
as HIV prevalence and other health indicators, is equal to or even
higher than in rural areas. These disparities are often not reflected
in national statistics, which mask the deprivation experienced in
poor urban neighbourhoods. Inequality in access to services,
housing, land, education, health care and employment opportu-
nities within cities have socio-economic, environmental and polit-
ical repercussions, including rising violence, urban unrest, envi-
ronmental degradation and underemployment, which threaten to
diminish any gains in income and poverty reduction.
This edition of the State of the World’s Cities Report provides
an overview of a range of issues that link cities, slums and the
Millennium Development Goals. It makes clear that the glob-
al fight against poverty – encapsulated in the Millennium
Development Goals – is heavily dependent on how well cities
perform. The Report highlights three inter-related issues:
• The Millennium Development Goals provide an apt frame-
work for linking the opportunities provided by cities with
improved quality of life;
• The achievement of the Goals depends on governments'
capacity to speed up progress in reducing urban poverty and
inequality and in reversing current trends in slum formation;
• Improving the living conditions of slum dwellers (housing,
tenure, infrastructure and access to basic services) will auto-
matically have a positive impact on the attainment of most
of the Goals and their related targets.
■ Where We Live Matters: The Social and
Health Costs of Living in a Slum
For as long as governments have been monitoring the human
development performance of their countries, achievements in
various sectors have tended to focus on only two geographical
areas: rural and urban. In general, statistics show that urban
populations are better off than those living in villages: they
tend to enjoy more access to services and generally perform well
on a range of human development indicators, including life
expectancy and literacy. However, evidence suggests that in
many developing countries, urban poverty is becoming as
severe and as dehumanizing as rural poverty. This Report pres-
ents for the first time data disaggregated at urban, rural, slum
and non-slum levels. The findings show remarkable similarities
between slums and rural areas:
• In low-income countries, such as Bangladesh, Ethiopia,
Haiti, India, Nepal and Niger, 4 out of every 10 slum chil-
dren are malnourished, a rate that is comparable to rural
areas of those countries.
• Likewise, in some cities, such as Khartoum and Nairobi,
the prevalence of diarrhoea is much higher among slum
children than among rural children. In slums, child deaths
are attributed not so much to lack of immunization against
measles, but inadequate living conditions, such as lack of
access to water and sanitation or indoor air pollution,
which lead to water-borne and respiratory illnesses among
children.
• Malnutrition and hunger in slums is almost the same as in
villages in some countries. In India, for instance, slum
dwellers suffer slightly more from malnutrition than the
rural population of the country.
• Recent data on HIV/AIDS shows that in various sub-
Saharan African countries, HIV prevalence is significantly
higher in urban areas than in rural areas, and is also high-
er in slums than in non-slum urban areas. Moreover, slum
women are particularly at risk, with HIV prevalence rates
that are higher than that of both men and rural women.
• Age pyramids for slum and rural populations in several
countries show similar patterns: both groups tend to be
younger and generally die sooner than non-slum urban
populations, which tend to have the lowest child mortali-
ty rates and the highest life expectancy rates.
The above examples show that slum populations are not
benefiting from the advantages and opportunities offered by
cities. Studies have also shown that children living in a slum
within a city are more likely to die from pneumonia, diar-
rhoea, malaria, measles or HIV/AIDS than those living in a
non-slum area within the same city; many of these diseases are
the result of poor living conditions prevalent in slums rather
than the absence of immunization coverage or lack of health
facilities. In many cases, poverty, poor sanitation and indoor
air pollution make children and women living in slums more
vulnerable to respiratory illnesses and other infectious dis-
eases than their rural counterparts. For many slum dwellers,
overcrowding, housing located in hazardous areas and the
threat of eviction affects other livelihood issues, such as
employment. Some studies have also found a strong correla-
tion between where people live and their chances of finding a
job. One such study in France showed that job applicants
residing in poor neighbourhoods were less likely to be called
for interviews than those who lived in middle- or high-
income neighbourhoods. Another study in Rio de Janeiro
found that living in a favela (slum) was a bigger barrier to
gaining employment than being dark skinned or female, a
finding that confirms that “where we live matters” when it
comes to health, education and employment.
These findings reveal “a tale of two cities within one city”.
Thus, policymakers, governments, development practitioners
and funding agencies should no longer see the city as one
homogenous entity. Slums are not only a manifestation of
poor housing standards, lack of basic services and denial of
human rights, they are also a symptom of dysfunctional urban
societies where inequalities are not only tolerated, but allowed
to fester.
v
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
This Report unfolds a new urban reality that shows how
poor living conditions impact the world’s slum dwellers: slum
dwellers die earlier, experience more hunger, have less educa-
tion, have fewer chances of employment in the formal sector
and suffer more from ill-health than the rest of the inhabi-
tants of cities.
The international community cannot afford to ignore
slum dwellers because, after rural populations, they represent
the second largest target group for development interven-
tions – and their size is set to grow as the developing world
becomes more urbanized. The Millennium Development
Goals thus have to target this disadvantaged and vulnerable
group of people; if they are ignored, it is very likely that the
Millennium Development Goals will not be achieved.
■ The State of the World’s Slums
The growth of slums in the last 15 years has been unprece-
dented. In 1990, there were nearly 715 million slum dwellers
in the world. By 2000 – when world leaders set the target of
improving the lives of at least 100 million slum dwellers by
2020 – the slum population had increased to 912 million.
Today, there are approximately 998 million slum dwellers in
the world. UN-HABITAT estimates that, if current trends con-
tinue, the slum population will reach 1.4 billion by 2020.
One out of every three city dwellers lives in slum conditions.
Some slums become less visible or more integrated into the
urban fabric as cities develop and as the incomes of slum
dwellers improve. Others become permanent features of urban
landscapes. Both types of slums have carved their way into
modern-day cities, making their mark as a distinct category of
human settlement that needs to be looked at over and above
the traditional rural-urban dichotomy.
Slum dwellers often live in difficult social and economic
conditions that manifest different forms of deprivation –
material, physical, social and political. Throughout this
Report, UN-HABITAT uses an operational definition of
slums – one with measurable indicators at household level.
Four of the five indicators measure physical expressions of
slum conditions: lack of water; lack of sanitation; overcrowd-
ing; and non-durable housing structures. These indicators –
known also as shelter deprivations – focus attention on the
circumstances that surround slum life, depicting deficiencies
and casting poverty as an attribute of the environments in
which slum dwellers live. The fifth indicator – security of
tenure – has to do with legality, which is not as easy to meas-
ure or monitor, as the status of slum dwellers often depends
on de facto or de jure rights – or lack of them. By knowing
how many slum dwellers there are in cities and what shelter
deprivations they suffer most from, it becomes possible to
design interventions that target the most vulnerable urban
populations.
Not all slums are homogeneous and not all slum dwellers
suffer from the same degree of deprivation. In this Report,
vi
UN-HABITAT presents an analysis of the degrees of shelter
deprivation in some selected countries and regions. This type
of information helps to connect monitoring information to
policy, making more rigorous and systematic the development
of programmes and interventions that are better attuned to
specific locations and situations.
The State of the World’s Cities Report 2006/7 provides an
overview of the state of the world’s slums with regards to the
five indicators. The following provides a summary of the main
findings.
Lack of durable housing
It is estimated that 133 million people living in cities of
the developing world lack durable housing. Non-durable or
non-permanent housing is more prevalent in some regions
than in others; over half the urban population living in
non-permanent houses resides in Asia, while Northern
Africa has the least numbers of people living in this kind of
housing. However, UN-HABITAT analysis shows that
global figures on housing durability are highly underesti-
mated due to the fact that durability is based primarily on
permanence of individual structures, not on location or
compliance with building codes. Moreover, estimates are
made taking into account only the nature of the floor mate-
rial, since information on roof and wall materials is collect-
ed in very few countries. For instance, figures indicate that
over 90 per cent of the world’s urban dwellings have perma-
nent floors, but when estimates are made combining floor,
roof and wall materials, this figure drops dramatically in
several countries. In Bolivia, for instance, when only floor
material is considered, 83.8 per cent of the urban popula-
tion is counted as having durable housing, but when wall
and roof materials are taken into account, this figure drops
to 27.7 per cent. Statistical analysis presented in this
Report shows that when more physical structure variables
are combined, the results provide a more realistic image of
housing durability.
Lack of sufficient living area
Overcrowding is a manifestation of housing inequality and
is also a hidden form of homelessness. In 2003, approximate-
ly 20 per cent of the developing world’s urban population –
401 million people – lived in houses that lacked sufficient liv-
ing area (with three or more people sharing a bedroom). Two-
thirds of the developing world’s urban population living in
overcrowded conditions resides in Asia; half of this group, or
156 million people, reside in Southern Asia. This Report
shows how living conditions, including overcrowding and
poor ventilation, are related to rates of illness, child mortality
and increase in negative social behaviors. It stresses that the
risk of disease transmission and multiple infections becomes
substantially higher as the number of people crowded into
small, poorly ventilated spaces increases. After presenting
overcrowding data by region, the Report highlights some of
the local variances of the definition.
Lack of access to improved water
Although official statistics reflect better water coverage in
urban areas than in rural areas, various surveys show that in
many cities, the quantity, quality and affordability of water in
low-income urban settlements falls short of acceptable stan-
dards. Improved water provision in the world’s urban areas was
reported to be as high as 95 per cent in 2002. This statistic,
however, presents an overly optimistic picture since “improved”
provision of water does not always mean that the provision is
safe, sufficient, affordable or easily accessible. For example, fur-
ther analysis reveals that getting water from a tap is a luxury
enjoyed by only two-third of the world’s urban population; less
than half of this group (46 per cent) have piped water within
their dwelling; 10 per cent rely on public taps, while 8 per cent
have access only to manually pumped water or protected wells.
Inter-regional differences indicate that Africa has the lowest
proportion (38.3 per cent) of urban households with access to
piped water, while the Latin American and Caribbean region
has the highest (89.3 per cent). Sometimes, even when water is
available, it may not be affordable or safe to drink. In Addis
Ababa, Ethiopia, a UN-HABITAT survey showed that the pro-
portion of low-income urban residents with access to water
supply dropped to 21 per cent from 89 per cent when the oper-
ational definition of “access” included variables such as cost and
quality. Poor access to water in urban areas has a direct bearing
on rates of water-borne or water-related diseases in urban areas,
a phenomenon that is explored in some depth in the latter part
of the Report.
Lack of access to improved sanitation
Over 25 per cent of the developing world’s urban popula-
tion – or 560 million city residents – lack adequate sanitation.
Asia alone accounts for over 70 per cent of this group, main-
ly because of the large populations of China and India; in
2000, sanitation coverage in Chinese cities was reported to be
approximately 33 per cent. UN-HABITAT analysis shows
that while cities in South-Eastern Asia and Southern Asia
have made significant progress in recent years to improve san-
itation coverage in urban areas, access lags far behind in sub-
Saharan Africa and Eastern Asia, where 45 per cent and 31
per cent of the urban population still lacks access to improved
sanitation, respectively. However, some countries in Southern
Asia have extremely low coverage, notably Afghanistan, where
only 16 per cent of the urban population has access to a prop-
er toilet. Lack of access to an adequate toilet not only violates
the dignity of the urban poor, but also affects their health.
Every year, hundreds of thousands of people die as a result of
living conditions made unhealthy by lack of clean water and
sanitation. The number of deaths attributable to poor sanita-
tion and hygiene alone may be as high as 1.6 million per year
– five times as many people who died in the 2004 Indian
Ocean tsunami. A disproportionate share of the labour and
health burden of inadequate sanitation falls on women, who
have to wait for long periods to gain access to public toilets or
have to bear the indignity of defecating in the open.
Inadequate sanitation is therefore something of a “silent
tsunami” causing waves of illness and death, especially among
women and children. As this Report shows, mortality rates
are quite often linked to whether or not children or their
mothers have access to adequate sanitation facilities; in the
city of Fortaleza in Brazil, for instance, child mortality rates
dropped dramatically when sanitation coverage increased.
Lack of secure tenure
Mass evictions of slum and squatter settlements in various
cities in recent years suggest that security of tenure is becom-
ing increasingly precarious, particularly in cities of sub-
Saharan Africa and Asia, where evictions are often carried out
to make room for large-scale infrastructure or city “beautifi-
cation” programmes. A global survey in 60 countries found
that 6.7 million people had been evicted from their homes
between 2000 and 2002, compared with 4.2 million in the
previous two years. Many of these evictions were carried out
without legal notice or without following due process.
Improving the tenure of urban households could go a long
way in preventing evictions, but operationalizing security of
tenure for the purpose of global monitoring remains difficult.
At present, it is neither possible to obtain household-level
data on secure tenure in most countries, nor to produce glob-
al comparative data on various institutional aspects of secure
tenure, as data on secure tenure is not regularly collected by
censuses or household surveys . However, non- empirical
information suggests that between 30 per cent and 50 per
cent of urban residents in the developing world lack security
of tenure. Although home ownership is regarded as the most
secure form of tenure, evidence from around the world also
suggests that ownership is not the norm in both the devel-
oped and the developing world, and is not the only means to
achieve tenure security. In fact, informal – or illegal – growth
has become the most common form of housing production in
the developing world, where gaining access to housing
through legal channels is the exception rather than the rule
for the majority of urban poor households. UN-HABITAT
and its partners are currently working on the preparation of a
global monitoring system that could in the future provide a
framework to assist governments at local and national levels
to produce estimates on how many people have secure tenure,
using an agreed-upon methodology in terms of definitions,
indicators and variables.
■ 30 Years of Shaping the “Habitat”Agenda:
Policies and Practices That Have Worked
Since the first UN Conference on Human Settlements
(Habitat I) took place in Vancouver in 1976, governments
and the international community have adopted and imple-
mented a range of human settlements policies and pro-
grammes with mixed results. Many programmes were unsuc-
cessful; others, while successful at the pilot stages, could not
vii
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
be scaled up and remained small “islands of success” that did
not have a significant impact on urban poverty levels or slum
growth rates. Few interventions had an economic or social
impact on urban poor populations.
Getting urban poverty on the development agenda has been
a struggle in the last thirty years. Silence or neglect have char-
acterized most policy responses. However, with the adoption
of the Millennium Declaration in 2000, urban poverty is now
being brought to the centre stage of the global development
agenda. As part of its mandate to assess the performance of
countries on Millennium Development Goal 7, target 11 – to
improve the lives of at least 100 million slum dwellers by
2020 – UN-HABITAT built a broad architecture for global
monitoring and reporting. As part of this process, the organ-
ization has evaluated the performance of more than 100
countries to see if they were “on track”, “stabilizing”, “at risk”
or “off track” vis-à-vis the slum target. Three criteria were
used to rate countries: annual slum growth rate; slum per-
centage; and slum population.
Analysis of the results revealed some interesting findings:
countries that had successfully reduced slum growth rates,
slum proportions and slum populations in the last 15 years
shared many attributes: their governments had shown long-
term political commitment to slum upgrading and preven-
tion; many had undertaken progressive pro-poor land and
housing reforms to improve the tenure status of slum
dwellers or to improve their access to basic services; most
used domestic resources to scale up slum improvements and
prevent future slum growth; and a significant number had
put in place policies that emphasized equity in an environ-
ment of economic growth. In many countries, improvements
in just one sector, such as sanitation, had a significant impact
on slum reduction, particularly in cities where inhabitants
suffered from only one or two shelter deprivations.
Another major finding of this analysis of country per-
formance on the slum target showed that those countries
doing well in managing slum growth had highly centralized
systems and structures of governance; even in cases where
decentralized systems existed, policy actions for slum pre-
vention and upgrading were implemented through central-
ized interventions. This was possible because central gov-
ernments –having command and control– could put in
place measures and resources to ensure cohesiveness in the
design and implementation of slum upgrading projects.
Central governments had the capacity to put forward legis-
lation and pro-poor policy reforms to tackle basic shelter
deprivations – reforms that require political support at the
national level before being filtered downward to local levels
of government. These central governments have been able
to set up the institutional arrangements, allocate important
budgets, and execute projects to effectively meet their tar-
gets and commitments. In countries such as Brazil, Egypt,
Mexico, South Africa, Thailand and Tunisia, implementa-
tion of inclusive policies, land reforms, regularization pro-
grammes and commitment to improve the lives of the
viii
urban poor by the top leadership were key to the success of
slum upgrading or prevention programmes. These coun-
tries developed either specific slum upgrading and preven-
tion policies or have integrated slum upgrading and preven-
tion as part of broader poverty reduction policies and pro-
grammes. They have done this not only to respond to social
imperatives, but also to promote national economic devel-
opment. Central governments in these countries, among
others, have played a critical role, not just in the physical
improvement of slums, but also in ensuring that invest-
Acknowledgements
ments are made in other sectors as well, such as education,
health, sanitation and transport, which have benefited slum Report Team
communities. Overall coordination: Don Okpala
This perhaps is a prelude to a change in governance para- Director: Nefise Bazoglu
digms, in which central governments and local authorities Task Manager: Eduardo López Moreno
would develop a more coordinated approach in the develop- Statistical Adviser: Gora Mboup
Editor: Rasna Warah
ment and implementation of policies, with central govern-
ments taking the lead in urban poverty reduction programmes
as they would have the power and authority to institute pro-
Principal Authors
poor reforms and the mandate and ability to allocate resources
Eduardo López Moreno and Rasna Warah
to various priority sectors. On the other hand, local authorities
would be able to locally coordinate operational actions bring-
ing together different actors. Contributors
This Report also clearly shows that not all countries strug- Nefise Bazoglu, Tanzib Chowdhury and Gora Mboup
gling to cope with high slum growth rates have shied away from
committing to change. Some sub-Saharan African countries,
namely Burkina Faso, Senegal and Tanzania, have in recent Additional input to individual chapters
years shown promising signs of growing political support for Cecilia Andersson, Christine Auclair, Francis Dodoo, Alex Ezeh, Anna Alvazzi del Frate,
slum upgrading and prevention that includes reforms in poli- Meg Holden, Asa Jonsson, Sunita Kapila, Wendy Mendes, Jaana Mioch, Luc Mougeot,
cies governing land and housing. Shipra Narang, Karen A. Stanecki, Darcy Varney, Ananda Weliwita and Eliya Zulu
Some low- or middle-income countries that are starting to
stabilize or reverse slum growth rates, including Colombia, Statistical Annex: Gora Mboup and Philip Mukungu
El Salvador, Philippines, Indonesia, Myanmar and Sri Graphs: Natsuo Ito
Lanka, did not wait to achieve important milestones in eco- Maps: Iris Knabe and Martin Raithelhuber
Design and Layout: Mike Jones
nomic growth in order to address slums. These countries
Editorial Assistance: Darcy Varney
have managed to prevent slum formation by anticipating Research Assistance: Martha Mathenge and Raymond Otieno
and planning for growing urban populations – by expand-
ing economic and employment opportunities for the urban
poor, by investing in low-cost, affordable housing for the UN-HABITAT Advisory and Technical Support
most vulnerable groups and by instituting pro-poor reforms Alioune Badiane, Daniel Biau, William Cobbett, Szilard Fricska, Anne Klen, Dinesh Mehta,
and policies that have had a positive impact on low-income Jane Nyakairu, Alberto Paranhos, Roman Rollnick, Sharad Shankardass, Wandia Seaforth,
people’s access to services. These countries give hope and Farouk Tebbal, Raf Tuts, Francisco Vasquez, Satyanarayana Vejella, Chris Williams and
direction to other low-income countries by showing that it Habitat Programme Managers in selected countries.
is possible to prevent slum formation with the right policies
and practices.
What comes out clearly in this Report is that slum formation International Advisory Board
is neither inevitable nor acceptable. “Running the poor out of Jo Beall, Joep Bijlmer, Andrew Boraine, Edesio Fernandes, Ilona Kickbusch, Susan
town” – through evictions or discriminatory practices – is not Loughhead, Miloon Kothari, Patricia L. McCarney, Luc Mougeot, Kalpana Sharma and
the answer: rather, helping the poor to become more integrat- Molly O’Meara Sheehan
ed into the fabric of urban society is the only long-lasting and
sustainable solution to the growing urbanization of poverty.
Ultimately, as the developing world becomes more urban and
as the locus of poverty shifts to cities, the battle to achieve the
Millennium Development Goals will have to be waged in the
world’s slums.
Contents
Part 1: Cities, Slums and the Millennium Development Goals
1.1 ‘City-zens’ of the World: Urban Trends in the 21st century ......................6
1.2 Putting Slums on the Map:
A Global and Regional Overview..............................................................18
1.3 How Well is Your Country Performing on the Slum Target?
A Global Scorecard ..................................................................................40
1.4 The Struggle to Achieve the Millennium Development Goals
will be Won or Lost in Cities ....................................................................48
Boxes
Defining “Urban”..................................................................................................................................................................7
What is a slum? ................................................................................................................................................................21
Slums: the shelter dimension of urban poverty ................................................................................................................28
Defining and monitoring slums: seeing beyond the stereotypes ....................................................................................29
Woman-headed households in cities ................................................................................................................................30
Age pyramids for slum and non-slum populations in Brazil and South Africa ................................................................31
Tables
1.2.1 Population of slum areas at mid-year, by region; 1990, 2001, 2005 and annual slum growth rate ....................18
1.2.2 Urban and slum growth rates by region ................................................................................................................20
1.2.3 Proportion of slum households in developing regions by number of shelter deprivations, 2001 ........................35
Figures
1.1.1 Proportion of urban population by region, 1950-2030 ............................................................................................7
1.1.2 Rural and urban population by region in 2005 and 2030 ........................................................................................8
1.1.3 Population by city size..............................................................................................................................................9
1.1.4 Urban growth in the world’s largest cities, 1950-2020 ........................................................................................10
1.2.1 Slum populations, 1990-2020 ................................................................................................................................21
1.2.2 Proportion of woman-headed households in selected countries..........................................................................30
1.2.3 Proportion of slum households in developing regions by number of shelter deprivations, 2001 ........................35
1.2.4 Magnitude of slums in 2020 under three scenarios..............................................................................................36
1.2.5 The rise and rise of slums, 1990-2020 ..................................................................................................................37
1.3.1 A global scorecard on slums..................................................................................................................................43
1.4.1 Human development is closely related to levels of urbanization ........................................................................48
1.4.2 The annual growth rate of cities and slums..........................................................................................................51
Maps
1 Urban population and slum proportion in African countries, 2001 ......................................................................23
2 Urban population and slum proportion in Asian countries, 2001 ........................................................................25
3 Urban population and slum proportion in countries of Latin America and the Caribbean, 2001 ........................33
4-6 A look a country performance in developing regions............................................................................................42
Cities + + = 8
Shelter People
Part 2: The State of the World’s Slums
2.1 Neither Brick nor Mortar: Non-Durable Housing in Cities ......................60
2.2 Not Enough Room: Overcrowding in Urban Households..........................70
2.3 Safe Drinking Water in Cities ..................................................................76
2.4 The Silent Tsunami:
The High Price of Inadequate Sanitation in Urban Areas........................84
2.5 Owners without Titles:
Security of Tenure in Cities of the Developing World..............................94
Boxes
Affordability, accessibility and durable houses ................................................................................................................62
Defining durability..............................................................................................................................................................65
Beyond culture: defining overcrowding ............................................................................................................................71
“Improved”or “adequate” access to water: definitions and issues of measurement......................................................78
Improved sanitation: a basic principle ..............................................................................................................................85
Defining sanitation: to estimate or to underestimate, that is the question ....................................................................86
Monitoring secure tenure ..................................................................................................................................................96
Tables
2.1.1 Housing durability, based on floor, roof and wall materials, in the urban areas of 16
selected countries, 2001 ........................................................................................................................................63
2.1.2 Finished floor coverage among urban population by region, 2003 ......................................................................64
2.2.1 Sufficient living area coverage among urban population by region, 2003 ..........................................................72
2.3.1 Improved drinking water coverage among urban population by region, 2003 ....................................................79
2.3.2 Cities making rapid progress in drinking water coverage, 1999-2003 ................................................................81
2.4.1 Improved sanitation coverage among urban population by region, 2003 ............................................................88
Figures
2.1.1 Distribution of the urban population lacking finished floor materials in developing regions, 2003....................64
2.1.2 Proportion of urban households with finished main floor materials, by region, 1990 and 2003 ........................67
2.2.1 Distribution of urban population lacking sufficient living area, 2003 ..................................................................72
2.2.2 Proportion of urban households with sufficient living area ..................................................................................74
2.2.3 Proportion of urban population with access to sufficient living area in selected cities, 2003............................74
2.3.1 Water expense as a fraction of household income in Addis Ababa, 2003 ..........................................................77
2.3.2 Distribution of urban population lacking improved drinking water by region, 2003 ............................................79
2.3.3 Access to piped water in selected cities in sub-Saharan Africa, 2003................................................................80
2.3.4 Access to water decreases dramatically when cost and quality are considered: the case of Addis Ababa ......83
2.4.1 Distribution of urban population lacking improved sanitation by region, 2003....................................................88
2.4.2 Proportion of urban households with access to improved sanitation ..................................................................90
2.4.3 Cities with low sanitation coverage:
proportion of urban population with access to improved sanitation, 2003..........................................................90
2.5.1 Percentage of urban households with adequate housing in Brazil, 1992-2003 ..................................................95
Maps
7 Proportion of households without finished main floor material, 2003 ................................................................63
9
Proportion of households lacking sufficient living area, 2003 ..............................................................................73
Proportion of households without access to improved water, 2003 ....................................................................79
10 Proportion of households without access to improved sanitation, 2003..............................................................89
Part 3: Where We Live Matters Part 4: Policies and Practices That Have Worked
3.1 The Social and Health Costs of Living in a Slum ..................................104 4.1 Milestones in the Evolution of
3.2 Hunger: The Invisible Crisis in Cities......................................................106 Human Settlements Policies 1976-2006 ................................................156
3.3 The Urban Poor Die Young......................................................................110 4.2 Countries Taking Slums Seriously ..........................................................162
3.4 HIV/AIDS and Urban Poverty ..................................................................116 4.3 Pro-poor Reforms on Slum Upgrading and Prevention ..........................166
3.5 Education and Youth Employment: 4.4 Governing from the Bottom, Governing from the Top,
Debunking Some Myths about the “Urban Advantage” ........................122 Connecting the Two ..............................................................................170
3.6 Cities: The Front Lines in the Battle for Sustainability ..........................130 4.5 Time for Bold Action:
3.7 Double Jeopardy: Scaling up Improvements Today, Preventing Slums Tomorrow ............176
The Impact of Conflict and Natural Disaster on Cities ..........................136 4.6 Is the International Community Ready to Keep the Promise? ..............180
3.8 Urban Insecurity: New Threats, Old Fears ............................................144
Boxes
Boxes Costing the slum target ..................................................................................................................................................164
Extreme deprivation in poor urban settings in sub-Saharan Africa................................................................................120 Improving urban planning and monitoring in the city of Aleppo ....................................................................................167
Renewable energy sources increase urban sustainability..............................................................................................134
Urban crime trends ..........................................................................................................................................................150
Figures
4.1.1 Slum population and urban population growth in the world (1976-2006)..........................................................157
Figures 4.1.2 GDP per capita by income groups 1975-2004 ....................................................................................................158
3.2.1 Slum incidence and proportion of underweight children in selected countries ................................................108
3.3.1 Under-five mortality (deaths per 1000 births) by type of residence in selected countries ................................111
3.3.2 Under-five mortality (deaths per 1000) births by type of residence in selected cities ......................................111
3.3.3 Diarrhoea prevalence among children under five years in selected countries ..................................................113
3.3.4 Proportion of underweight children under age five in selected African cities ..................................................114
3.3.5 Proportion of children under age five with Acute Respiratory Infections
in selected Latin American and Asian cities ......................................................................................................115
3.4.1 HIV prevalence among men and women aged 15-49 in urban and rural areas
3.5.1
in selected sub-Saharan African countries 2000-2004 ......................................................................................116
Net enrolment rate (primary) by type of residence in selected countries ..........................................................124
3.5.2 Proportion of women aged 15-24 who stopped going to school because of
inability to pay school fees ..................................................................................................................................126
3.5.3 Percentage of young women and men working in the informal sector in selected African countries ..............128
3.5.4 Percentage of young women and men who have family responsibilities in selected countries ......................128
3.5.5 Percentage of young women and men working in the informal sector in selected Asian countries ................129
City Stories Statements
China’s rising cities......................................................................................................................................................16 Peter Hall: Why some cities flourish while others languish ......................................................................................15
Mumbai’s quest for ‘world city’ status ......................................................................................................................26 Scott Leckie: The slum target is not in line with housing rights................................................................................38
Europe’s forgotten Roma community ..........................................................................................................................68
Miloon Kothari: Water and sanitation: only a human rights approach will do..........................................................82
Sanitation: A women’s issue ......................................................................................................................................92
Evictions fail to address the root cause of urban poverty in Zimbabwe..................................................................100 Lena Sommestad: Stronger action is needed to achieve the sanitation target ........................................................93
Agrocities: combating hunger in urban areas ..........................................................................................................109 Luis Felipe Cabrales Barajos: Gated communities are not the solution to urban security......................................148
Slum conditions increase risk of HIV infection in Nairobi........................................................................................119 Stephen Graham: The urbanization of political violence..........................................................................................152
Immigrants in Paris: dreams go up in flames ..........................................................................................................125
Surviving anarchy: Somalia’s experience..................................................................................................................139
New Orleans: poor residents suffer deepest impact of Hurricane Katrina..............................................................142
Elisabeth Gateau: Localizing the Millennium Development Goals ..........................................................................161
Tony Hill: Civil society and the urban agenda ..........................................................................................................172
Vancouver: The world’s most liveable city combines multiculturalism with environmental sustainability ............160 Antoine Heuty and Sanjay Reddy: Technical fixes are not always the solution......................................................175
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

1
Part One

Cities, Slums and the

Millennium Development Goals

This Part highlights the major urbanization trends

in the world and portrays a global and regional
overview of slums. It also presents the first
findings of country performance on the Millennium
Development Goals’ slum target using a new global
scorecard developed by UN-HABITAT. The Part
concludes by emphasizing the importance of
implementing the Millennium Development Goals
at the city level, and more importantly, in slums. A
table linking cities and slums to the Millennium
Development Goals is also presented.
Bangkok, Thailand JEARANAIKUL/UNEP/STILL PICTURES
 S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
1.1 ‘City-zens’ of the World: Urban Trends in the 21st Century
Havana, Cuba RASNA WARAH
■ 2007: The dawn of the urban millennium
T
he year 2007 will mark a turning point in
human history: the world’s urban population
will for the first time equal the world’s rural
population. Although it is difficult to predict on
which day or month this radical transformation
will occur, what is certain is that this milestone will herald the
advent of a new urban millennium: a time when one out of
every two people on the planet will be a “city-zen”.
Cities, whether small municipalities of 2,000 inhabitants or
massive agglomerations of 10 million people or more, are
becoming a widespread phenomenon. The global urban pop-
ulation has quadrupled since 1950, and cities of the develop-
ing world now account for over 90 per cent of the world’s
urban growth.
In 2005, the world’s urban population was 3.17 billion out of
world total of 6.45 billion. Current trends predict the number
6 7
of urban dwellers will keep rising, reaching almost 5 billion by
2030. Between 2005 and 2030, the world’s urban population is
expected to grow at an average annual rate of 1.78 per cent,
almost twice the growth rate of the world’s total population. As
more and more people occupy cities, the population of rural set-
tlements around the globe will begin to contract after 2015,
decreasing at an average annual rate of -0.32 through 2030 – a
decrease of more than 155 million people over 15 years.1
■ Asia and Africa will host the largest urban
populations
Whereas Europe, North America and Latin America experi-
enced intense urbanization – the increased concentration of
people in cities rather than in rural areas – and rapid urban
growth through the mid-20th century, the trend has now
shifted to the developing regions of Asia and Africa. In-migra-
tion, reclassification and natural population increase are con-
tributing to a rapid urban transformation of these regions.
Annual urban growth rates are highest in sub-Saharan Africa
(4.58 per cent), followed by South-Eastern Asia (3.82 per
cent), Eastern Asia (3.39 per cent), Western Asia (2.96 per
cent), Southern Asia (2.89 per cent) and Northern Africa
(2.48 per cent). The developed world’s cities are growing at a
slower pace, averaging 0.75 per cent a year.
Latin America is the most urbanized region in the developing
world, with 77 per cent of its population – 433 million people
– living in cities.2 The urbanization of Latin America has yet to
reach its peak; by 2015, it is predicted that 81 per cent of its
population will reside in urban areas. In terms of sheer numbers,
however, Asia has the largest urban population (with more than
1.5 billion people inhabiting its cities) even though slightly less
than 40 per cent of its population is urbanized. The total popu-
lation of cities in the developing regions of the world already
exceeds that of cities in all of the developed regions (by 1.3 bil-
lion people). If predictions prove accurate, by 2030, nearly 4 bil-
lion people – 80 per cent of the world’s urban dwellers – will live
in cities of the developing world.
Asia and Africa will continue to dominate global urban
growth through 2030. Currently the least urbanized regions in
the world, with 39.9 per cent and 39.7 per cent of their pop-
ulations living in cities in 2005, respectively, by 2030, both
regions will become predominantly urban, Asia with 54.5 per
cent of its population living in cities, and Africa with 53.5 per
cent of its population urban. Asia alone will account for more
than half the world’s urban population (2.66 billion out of a
global urban population of 4.94 billion); and the urban pop-
ulation of Africa (748 million) will by 2030 be larger than the
total population of Europe at that time (685 million).
Defining “Urban”
The United Nations defines an urban agglom-
eration as the built-up or densely populated
area containing the city proper, suburbs and con-
tinuously settled commuter areas. It may be
smaller or larger than a metropolitan area; it
may also comprise the city proper and its subur-
ban fringe or thickly settled adjoining territory.
A metropolitan area is the set of formal local
government areas that normally comprise the
urban area as a whole and its primary com-
muter areas.
A city proper is the single political jurisdiction
that contains the historical city centre.
Sources: United Nations: Principles and Recommendations for Population and Housing Censuses (1998) and World Urbanization Prospects: The 2003 Revision
■ Small and intermediate cities will absorb most
urban growth
Small cities with less than 500,000 inhabitants and inter-
mediate cities with between 1 and 5 million inhabitants, not
megacities (defined as cities with 10 million or more people),
will continue to absorb most of the urban population around
the world well into the future. More than 53 per cent of the
world’s urban population lives in cities of fewer than 500,000
inhabitants, and another 22 per cent of the global urban pop-
ulation lives in cities of 1 to 5 million inhabitants. These
FIGURE 1.1.1 PROPORTION OF URBAN POPULATION BY REGION, 1950-2030
100
80
Percentage of urban population 60
50
40
20
0
1950 1955 1960 1965 1970
Africa
North America
Source: United Nations, World Urbanization Prospects: The 2003 Revision.
However, an analysis of countries shows that • 25 countries specify economic character-
different criteria and methods are currently istics as significant, though not exclusive,
being used by governments to define “urban”: in defining cities – typically, the proportion
of the labour force employed in non-agri-
• 105 countries base their urban data on cultural activities.
administrative criteria, limiting it to the • 18 countries count the availability of
boundaries of state or provincial capitals, urban infrastructure in their definitions,
municipalities or other local jurisdictions; 83 including the presence of paved streets,
use this as their sole method of distinguishing water supply systems, sewerage systems,
urban from rural. or electric lighting.
• 100 countries define cities by population • 25 countries provide no definition of
size or population density, with minimum “urban” at all.
concentrations ranging broadly, from 200 to • 6 countries regard their entire popula-
50,000 inhabitants; 57 use this as their sole tions as urban.
urban criterion.
cities are significant sites of social and economic activity,
often serving as centres of trade and destinations for rural
migrants.3 They are often the first places where the social
urban transformation of families and individuals occurs; by
offering economic linkages between rural and urban environ-
ments, they can provide a “first step” out of poverty for
impoverished rural populations and a gateway to opportuni-
ties in larger cities. In Eastern Africa, South-Eastern Asia, the
Caribbean and Europe, cities of fewer than 500,000 are par-
ticularly prevalent, hosting approximately two-thirds of those
regions’ urban residents.
1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030
Asia Latin America and the Caribbean Oceania
Europe World
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

Most of the world’s urban population will continue to live
in small cities over the next decade, but intermediate cities are
predicted to grow at a faster rate: between 2000 and 2015,
cities of fewer than 500,000 will likely increase their popula-
tions by 23 per cent, while cities of 1 to 5 million are predict-
ed to increase their populations by 27 per cent. By 2015, Asia
will have gained 37 cities of 1 to 5 million people, rising to a
total of 253; Africa will have gained 20, totaling 59; and Latin
America and the Caribbean will have gained 16, rising to a
total of 65.
■ The emergence of the “metacity”
Although “megacities” of more than 10 million inhabitants
have been around since the 1950s, when New York and Tokyo
were the largest cities in the world, “metacities”4 – massive
tion, and most have grown at the relatively slow rate of about
1.5 per cent annually. Although new research techniques that
combine population statistics with satellite imagery reveal that
these huge urban agglomerations may already be home to 7 per
cent of the world’s population, they still represent just a small
minority of cities worldwide.6
The first metacity came into being in the mid-1960s when
Tokyo’s population crossed the 20 million inhabitant threshold.
Tokyo continues to be the only metacity in the world today,
with a population in excess of 35 million people – more than
the total population of Canada. In less than a decade, however,
Mumbai, Delhi and Mexico City will have joined the league of
metacities, closely followed by São Paulo, New York, Dhaka,
Jakarta and Lagos, each with more than 17 million inhabitants.
By 2020, all of these cities are expected to attain metacity sta-
tus.7 Lagos is experiencing an exceptional growth rate – more
than 5 per cent per year through 2005 – and is expected to con-

Street Scene, Morocco NATSUO ITO

conurbations of more than 20 million people – are now gaining
ground in Asia, Latin America and Africa.
Called “hypercities” by some,5 cities of more than 20 million
inhabitants represent a new type of settlement above and
beyond the scale of megacities. Spurred by economic develop-
ment and increased population, they gradually swallow up rural
areas, cities and towns, becoming multi-nuclear entities count-
ed as one. The world has never before known so many cities as
large as these metacity agglomerations. Many of these cities have
populations larger than entire countries; the population of
Greater Mumbai (which will soon achieve metacity status), for
instance, is already larger than the total population of Norway
and Sweden combined.
Contrary to common perception, however, the world’s largest
cities are home to only 4 per cent of the world’s total popula-
FIGURE 1.1.2 RURAL AND URBAN POPULATION BY REGION IN 2005 AND 2030
5,000
4,000
Population (millions)
tinue growing faster than the other largest cities of the world
through 2020.
On average, the largest cities grow more slowly than cities
of 5 to 10 million people, and they account for only about 9
per cent of the world’s urban population. The impact of the
largest cities on their regions, however, is great. Between 1950
and 2020, the New York-Newark metropolitan area is expect-
ed to have increased its population by only 40 per cent,
whereas Mumbai will have grown by 88 per cent, and Dhaka
will have grown to more than 50 times its size in 1950.
Throughout the past half-century, the largest cities of the
developing world have had to absorb astounding increases in
their urban populations while many cities in the developed
regions have grown considerably less, or, as in the case of
London, decreased in population.
FIGURE 1.1.3 POPULATION BY CITY SIZE
Africa is the least urbanized continent but by
2030 its urban population will exceed the total
population of Europe.
■ City-regions will require new forms of
More than 53 per cent of the world’s coordinated management for sustainability
urban population lives in cities of
fewer than 500,000 inhabitants By 2020, all but 4 of the world’s largest cities will be in
developing regions, 12 of them in Asia alone. While still few
in number, these metacities point to new forms of urban
planning and management, leading to the growth of city-
regions and “metropolitanization”.
As cities increase in size, metropolitanization is becoming a
progressively more dominant mode of urbanization, particu-
larly in megacities and newly emerging metacities.
4,000
Metropolitanization can take various forms: it may involve
3,500 densely settled regions in which villagers or people living in
suburbs commute to work in the nearby cities but where many
3,000
of the production and service activities are located in rural
Population (millions)
2,500 areas and suburbs; it may mean that a stagnating and declin-

3,000 ing population and economic base in the core of a city shifts
2,000
to nearby secondary cities; or it may refer to the development
Rural population 2005 (millions)
1,500 of inter-connected systems of cities that create city-regions
2,000 Urban population 2005 (millions) linked by manufacturing and other activities, such as the
1,000
Rural population 2030 (millions)
Hong Kong/Pearl River Delta region of China.8
500 Metropolitanization calls for new, innovative and more
1,000 Urban population 2030 (millions)
decentralized forms of governance. Already, many large cities
0
2000 2005 2010 2015 are decentralizing governance to the appropriate levels with
0 more municipalities and boroughs managing different parts of
Europe North Africa Asia Latin Oceania 10 million or more 1 to 5 million Fewer than 500,000 the city. This calls for better inter-municipal coordination,
America America and 5 to 10 million 500,000 to 1 million
the Caribbean more intermediate metropolitan levels of governance, more
civil society participation and more autonomy for various
Source: United Nations, World Urbanization Prospects: The 2003 Revision. Source: United Nations, World Urbanization Prospects: The 2003 Revision. parts of this new organism called the metacity.

8 9
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

The scale of environmental impact of metacities and megaci-
ties on their hinterlands is also significant and is likely to be a
cause for concern in coming decades. For instance, in China,
rapid economic growth and urbanization, combined with
inconsistent implementation of industrial emissions standards
institutions, but also generating related economic and civil
society activity in other cities. These “world cities” provide
economies of scale and access to resources of local and global
significance. Connectivity, economic production and cultural
innovation have long kept London, New York, Paris, and

Simit Seller, Turkey DARCY VARNEY

and increased use of motor vehicles, have had a negative impact
on the urban environment; the country is the largest producer
of greenhouse gases, after the United States, and hosts 16 of the
20 most polluted cities in the world. Managing environmental
sustainability, economic sustainability and socio-political sus-
tainability – the three pillars of sustainable urbanization – will
require more polycentric forms of governance, more environ-
mentally-friendly legislation and a regional approach to plan-
ning and management of human settlements.
■ Competition between “world cities” will
intensify
Cities are more than simply concentrations of people and
resources. As hubs of trade, culture, information and industry,
cities also articulate and mediate major functions of the global
economy. In developed countries, cities generate over 80 per
cent of national economic output, while in developing coun-
tries, urban economic activity contributes significantly to
national revenue, generating up to 40 per cent of gross domes-
tic product.9 Wealthy world cities are also increasingly operat-
ing like city-states and city-regions, independent of regional or
national mediation.10
Global urban economies are increasingly reliant on advanced
producer services for their income: advertising, finance, bank-
ing, insurance, law, management consultancy, and other serv-
ice-based businesses.
Today, several major cities play pivotal roles in global net-
works, not only producing goods and services and hosting
FIGURE 1.1.4 URBAN GROWTH IN THE WORLD’S LARGEST CITIES, 1950-2020
40,000
35,000
30,000
Population ('thousands)
Tokyo at the top of the world-city scale, closely followed by
cities such as Frankfurt, Hong Kong, Amsterdam, Singapore,
São Paulo and Shanghai, which are emerging as trend-setters
on the global financial scene. Other cities, such as Dubai and
Rotterdam, are becoming global transport hubs, while
Bangalore, Seattle and Silicon Valley have emerged as world
leaders in the area of information technology.
In the new urban millennium, world city status is beginning
to extend to several key cities in developing regions as well,
based on their new roles in the global economy and their capac- In developing countries, informal
ity for linking resources with populations in need. Cities such employment comprises one-half to
as Istanbul and Mumbai are already establishing the cultural three-quarters of non-agricultural
trends in their countries and regions, and this influence could employment. In many of these
cross international borders through films, literature and satel-
countries, the informal sector provides
lite television networks and entertainment. Large cities in the
developing world, including Nairobi, Addis Ababa and more employment opportunities than
Bangkok, among others, are increasingly bringing together the formal sector.
major national and international partners by hosting interna-
tional agencies and development partners, and offering avenues
for constructive peer exchange, mediation and diplomacy.
At the same time, cities are becoming more competitive with
each other. No longer are only the world’s most highly recog-
nized cities jockeying for the honour of hosting major interna- ■ The urban economy in the developing world and Northern Africa, it is 51 and 48 per cent, respectively. If
tional events and corporate headquarters; growing cities of the will be largely informal informal employment in the agricultural sector is also includ-
developing regions are also competing with each other to ed, the proportion of informal employment in the total labour
become important regional, corporate and development cen- In the developing world, there has been a trend toward force is even larger. It is common to attribute the high propor-
tres. The primary economic rival of India’s financial capital of “informalization” of the urban economy, with increasing shares tion of informal sector workers in these regions to migration of
Mumbai, for example, is Shanghai in neighbouring China, of incomes earned in unregulated employment.11 In many unskilled workers from rural areas to urban areas. However, in
which has similar global aspirations, but has a much smaller developing countries, particularly in Africa and Asia, the formal many regions, informal sector employment is the only option
proportion of its population living in poverty. sector has not been able to provide adequate jobs for rapidly available to skilled and educated people who are not absorbed
growing urban populations, leading to the proliferation of the by the formal labour market. In addition to its contribution to
urban informal sector. In Latin America and the Caribbean, 7 employment, the sector also contributes significantly to nation-
out of 10 new jobs in urban areas are created in the informal al economies in terms of income. The sector’s share of gross
sector.12 Two main processes have significantly contributed to domestic product (GDP) is approximately 41 per cent in sub-
the rise in urban informal activities. One is the failure of the Saharan Africa, 31 per cent in Asia, 29 per cent in Latin
formal sector to provide adequate jobs and income-generating America and 27 per cent in North Africa.15
opportunities for a rapidly growing urban population; the Women account for a disproportionately larger share of the
other is the growing tendency of the formal sector to contract informal labour force than men, particularly in sub-Saharan

25,000
services out to secondary labour markets, which are mainly in Africa and Asia. In developing countries as a whole, more than 60
20,000 the informal sector. per cent of women are engaged in informal employment in the
In developing countries, informal employment comprises non-agricultural sector. In sub-Saharan Africa, about 84 per cent
15,000 1950
one-half to three-quarters of non-agricultural employment. In of women are employed in non-agricultural informal activities
2005
10,000 many of these countries, the informal sector provides more compared with 63 per cent of men. In Latin America, 58 per cent
2015
employment opportunities than the formal sector. In sub- of women are engaged in non-agricultural informal sector activi-
5,000 2020 Saharan Africa, for instance, the informal sector accounts for ties compared with 48 per cent of men. Women’s participation in
0 about 78 per cent of all non-agricultural employment.13 In both the informal and formal sectors of the urban economy has
Kenya alone, there are an estimated 5.5 million informal sector had a positive impact on their social mobility and political
w rk-

on

ris

Ci co

Pa Sao

s
ba

lh

go
ky

ak
De
nd

Pa

i
Ne Yo
k

ty

o
um

ex
To

Dh

La
ar

ul
Lo

M
w

workers compared with only about 1.7 million wage-earners in involvement in urban affairs. A study in Dhaka, Bangladesh, for
M
Ne

Source: United Nations, World Urbanization Prospects: The 2003 Revision. formal establishments.14 In Asia, 65 per cent of all the non-agri- instance, found a positive correlation between women’s employ-
Note: Population in 2020 was estimated from population in 2010 and 2015 assuming that trends for these years remain the same. cultural employment is in the informal sector. In Latin America ment in factories and their level of political participation.16

10 11
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
■ Insecurity will be a growing concern in cities of
the developed world
Since the attacks on New York and Washington on 11
September 2001, cities of the developed world have become
increasingly concerned about their vulnerability to acts of ter-
rorism. These concerns have been magnified by recent attacks
on cities such as London, Madrid, Bali, New Delhi, Nairobi
and Dar es Salaam. Because of their dense populations and
intricate infrastructure, cities are deeply affected by attacks,
which exact heavy physical, psychological and financial costs. It
is estimated that the city of New York lost $110 billion in infra-
structure, buildings and jobs as a result of the 11 September
attacks.17
Although terrorism affects both developed and developing
countries, the former have in recent years put in place measures
to respond to the crisis by increasing their budgets for security
and surveillance apparatus and tightening immigration poli-
cies. In some cities, such as New York and London, this has
meant visible changes in urban form: loss of use of public
space, restriction of movement within the city, weakening of
popular participation and increased use of security and surveil-
lance equipment in strategic locations. Terrorism and insecuri-
ty create fear and change perceptions in cities, leading to situa-
tions in which “a bag is no longer a bag, but a bomb”. A 2003
report by the City of London Police found that almost one in
ten Londoners worried about the threat of terrorism on a daily
basis. London authorities are said to have increased police
patrols and installed more surveillance cameras in strategic
locations, making the city one of the most closely surveilled
spaces in the United Kingdom, and perhaps the world.18 Some
are concerned that the implementation of these measures could
mean that less money will be available for social services, such
as health and education, which could become a source of fric-
tion and conflict in the future.
While terrorism dominates the concerns of cities of the
developed world, most developing country cities are contend-
Cairo ©ALESSANDRO BOLIS. IMAGE FROM BIGSTOCKPHOTO.COM
12
ing with other forms of insecurity that threaten their lives and
livelihoods. The security of the urban poor, in particular, is
affected by their health status, which influences both their abil-
ity to work and their ability to escape poverty. The HIV/AIDS
pandemic has particular implications for urban security as it
leads to loss of household income, growth in the phenomenon
of orphaned street children and disintegration of the family
unit. Many urban poor families also face the constant threat of
eviction. Insecurity is exacerbated by insecure tenure with
respect to both housing and land.
■ Divided cities: Cities are, and will continue to
be, sites of extreme inequality
In recent years, an increasing number of countries have
opened up markets and expanded political freedoms. Others
have made impressive gains in economic growth. But democra-
cy and economic growth have not helped reduce inequalities in
much of the world; the wealthiest 20 per cent of the world’s
people account for 86 per cent of private consumption, while
the poorest account for just 1 per cent. In the past six years, 23
million more Latin Americans slipped into poverty, and most
African countries – with the exception of Botswana and Egypt
– are poorer today than they were in the 1970s. Despite
impressive economic growth rates in both China and India in
the last decade, these two countries have not been able to join
the ranks of “high-income” countries, and have been unable to
bridge the income gap between rural and urban populations
and between the urban rich and the urban poor.19
These inequalities manifest themselves most starkly in cities.
In many cities of both the developed and the developing world,
economic growth has not resulted in prosperity for all. On the
contrary, intra-city inequalities have risen as the gap between the
rich and the poor has widened. Although poverty remains a pri-
marily rural phenomenon, large sections of the population in
urban areas are suffering from extreme levels of deprivation.
One out every three city dwellers – nearly
one billion people – lives in a slum.
The highest levels of income inequality exist in Africa and
Latin America, the least and most urbanized developing regions,
respectively. This inequality is most stark in urban areas, and par-
ticularly in large cities. Although the proportion of poor people
in rural areas is larger than the proportion in urban areas, there
are more poor people living in Latin America’s cities than its rural
areas. In 1999, for example, only 77 million of the region’s 211
million poor lived in rural areas, while the remaining 134 million
lived in urban areas, although the proportion of rural poor was
much greater than that of the urban poor – at 64 per cent and
34 per cent, respectively. In Africa, the proportion of people liv-
ing in poverty in rural areas is 59 per cent, compared with 43 per
cent in cities, a gap that is likely to shrink in an environment of
economic decline.20 Sub-Saharan African countries have some of
the world’s highest levels of urban poverty, extending to more
than 50 per cent of the urban populations in Chad, Niger and
Sierra Leone. Countries of Northern Africa and Western Asia
have urban poverty levels near or below 20 per cent. In Asia,
India has the highest urban poverty levels, at 30 per cent.
In Latin America and the Caribbean, levels of urban poverty
vary widely, from 8 per cent of the urban population in
Colombia to 57 per cent in Honduras. In some countries
around the world – notably Nigeria, Egypt and Trinidad and
Tobago – urban and rural poverty percentages are almost
equal.21 However, income-based statistics should be viewed with
caution as the true extent of urban poverty is likely higher than
they suggest. The high cost of non-food items, such as transport,
health, education, and water in cities – and poor living condi-
tions, including inadequate housing and poor access to water
and sanitation – impact the ability of the urban poor to rise out
of poverty. When these items are included to measure poverty,
poverty estimates for urban areas are likely to rise significantly.
■ Slums are emerging as a dominant and distinct
type of settlement in cities of the developing
world
One out every three city dwellers – nearly one billion peo-
ple – lives in a slum. The vast majority of slums – more than
90 per cent – are located in cities of the developing world,
which are also absorbing most of the world’s urban growth.22
Urbanization has become virtually synonymous with slum
growth, especially in sub-Saharan Africa, Western Asia and
Southern Asia, where annual slum and urban growth rates
are almost identical. Annual slum and urban growth rates are
highest in sub-Saharan Africa, 4.53 per cent and 4.58 per
cent respectively, nearly twice those of Southern Asia, where
slum and urban growth rates are 2.2 per cent and 2.89 per
cent, respectively. In Western Asia, slums and cities are grow-
ing at a similar pace, 2.71 per cent and 2.96 per cent respec-
tively. Northern Africa is the only sub-region where slum
growth rates are declining, largely due to positive measures
taken by individual countries to address the plight of slum
dwellers. Eastern Asia, South-Eastern Asia and Latin
America and the Caribbean are also regions where annual
slum growth rates have not kept pace with annual urban
growth rates. Nonetheless, these regions continue to have
large numbers of their urban populations residing in slums.
Sub-Saharan Africa has the highest prevalence of slums in
the world – 71.8 per cent of its urban population lives in
slums – and in the last 15 years, the number of slum dwellers
in the region has almost doubled, from 101 million in 1990
to 199 million, in 2005. Given the high slum growth rate in
the sub-region, the number of slum dwellers is projected to
double by 2020, reaching nearly 400 million, and overtaking
the slum populations of both Southern Asia and Eastern
Asia, where slum populations are projected to rise to 385
million and 299 million, respectively. In terms of absolute
numbers, Asia still has the largest share of the world’s slum
13
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

PETER HALL
WHY SOME CITIES FLOURISH WHILE OTHERS LANGUISH
Endnotes
Charcoal vendor, Kibera, Nairobi HIROSHI SATO

There is a vital question for cities, but Very well, policymakers may say; what
1 All urban population statistics are drawn from the United Nations
the answer eludes urbanists: at a time are the lessons for the places now
Department of Economic and Social Affairs, Population Division, World
when the knowledge economy is competing to become the 21st century
Urbanization Prospects: The 2003 Revision, unless otherwise noted.
2 Of the more-developed regions, Australia and New Zealand have the
becoming all-pervasive, what makes equivalents? One point is clear: the
highest proportion of their populations living in cities: 91.6 per cent in cities innovative? Why do cities flour- candidates are no longer individual
2005. ish creatively, but then languish? Why cities but the megacity-regions of
3 Satterthwaite & Tacoli 2003. are Athens or Florence no longer lead- Southeast England, the Northeastern
4 Term coined by UN-HABITAT for cities with populations of more than 20 ing creative cities? Why have seaboard of the United States and the
million. Guangzhou and Shanghai taken the Yangtze River Delta in China, clustered
5 Davis, 2004. places of Manchester and Detroit? around global cities such as London,
6 The Center for International Earth Science Information Network 2005. How do a few cities, including London New York and Hong Kong. Here we
7 This prediction is based on a linear trend model, using the population and New York, manage to retain their find the 21st century equivalents of
growth between 2010 and 2015 as the baseline to extend the population edge? 18th century Manchester or 19th cen-
to 2020. tury Berlin. Centres of concentrated
8 UN-HABITAT/DFID 2002. Economists approach this question innovative power, now diffusing into
9 UN-HABITAT 2002c.
through regression equations to try to neighbouring cities and towns through
10 Taylor 2005.
identify the critically important ingre- networks of information exchange:
11 Cohen 2004.
dient in successful cities. But this may London to Reading and Milton Keynes,
12 Inter-American Development Bank 2004.
not provide a robust explanation for all Shanghai to Suzhou.

Sub-Saharan African countries
have some of the world’s highest
levels of urban poverty.
population; in 2005, the region housed more than half the
world’s slum dwellers, or 581 million people.
Slums in many cities are no longer just marginalized neigh-
bourhoods housing a relatively small proportion of the urban
population; in many cities, they are the dominant type of
human settlement, carving their way into the fabric of mod-
ern-day cities, and making their mark as a distinct category
of human settlement that now characterizes so many cities in
the developing world. Although slums do not directly denote
levels of urban poverty, their prevalence in a city can be an
indicator of urban inequality. UN-HABITAT projections
indicate that the number of slum dwellers in the world will
rise to 1.4 billion by 2020 if no remedial action is taken.
London MJS
13 International Labour Organization 2002.
14 Bindra 2005.
or most cities over time. There is
15 International Labour Organization 2002. another way, through economic and
16 Sachs 2005. social history: carefully dissecting the
17 Cohen 2004. sequence and combination of causes that facilitated the emergence of
18 Coaffee 2004. creative cities. The danger with this approach, too, is that it may fail
19 United Nations 2005a. to generate good general explanations.
20 Ibid.
21 World Bank 2002 estimates. Yet it need not. In my own work on creative cities, common themes
22 All slum data drawn from UN-HABITAT Global Urban Observatory. emerge. All were leading cities of their age economically; invariably
For a definition of “slum”, see Chapter 1.2. they were the centres of trading empires. They were at the economic
forefront, or near it. Thus, they became magnets for people with abil-
ity, who migrated from far corners of their far-flung empires. It is no
accident that key roles were played by outsiders: non-citizen Metics in
Periclean Athens, Jews in the Vienna of the early 1900s, artists like
Pablo Picasso in Paris shortly after. The immigrants considered them-
selves half inside, half outside the established societies in which they
lived. They became creative lightning rods of a sort, illuminating the
underlying tensions inside these societies. Occasionally, as in Berlin in
the 1920s, such tensions tore those societies apart. Generally, the
results were happier.
Technological innovation shows similar features – but also interesting
differences. Manchester in 1780, Detroit in 1910 and Silicon Valley in
1960 were upstart cities or city-regions, egalitarian places that wel-
comed new talent, stressing individual self-improvement and mutual
education. Their people engaged in extraordinary chains of innovation,
through networks that – paradoxically – were simultaneously compet-
itive and cooperative. There are amazing parallels between Lancashire
during the period between 1760 and 1830 and Silicon Valley since
1960: one innovation stimulated another, in long and complex chains.
Regions like these became creative because of an extraordinary
process of mutual learning and mutual stimulation.
The key is information. Knowledge is
the new production factor, and the
truly vital information is exchanged face-to-face, brain-to-brain,
through local networks in quite concentrated downtown areas, such
as the City of London or Downtown Manhattan. So the critical ques-
tion now concerns the pattern of diffusion and reconcentration, and
the limits to that process, within these multi-centred city-regions, set
by some critical time limit from the central city.
Information, the raw material of the knowledge economy, is manipu-
lated by workers in the advanced producer services, who generate
incomes that trigger a dazzling array of consumer services. Megacity-
regions compete in a winner-takes-all contest. The most successful
city-regions grow faster, reinforced by advantages in communications,
such as a major international airport or train hub. It is also no accident
that London and New York, the great English-speaking cities, repre-
sent the twin peaks of this global information economy.
Yet in the coming century, this may change. The 21st century will
clearly be the Asian century. China and India are already racing ahead,
aiming to regain the leading positions they occupied in past centuries.
Their resurgence will come through their great cities – Shanghai,
Beijing, Mumbai – and their surrounding megacity-regions. This will
be an ongoing East-West economic Olympic Games – but the stadi-
ums, where the prizes will be won, will be urban.
Sir Peter Hall is Professor of Planning at the Bartlett School of
Architecture and Planning, University College London. He is the author
and editor of over 30 books, including Cities of Tomorrow (1988) and
Cities in Civilization (1998).

14
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
China’s Rising Cities
China PHOTOS.COM
Recent economic reforms coupled with modernization
policies have improved the living conditions of millions of
people in China. In the last two decades, the world’s most
populous country has witnessed annual economic growth
rates of more than 9 per cent while the proportion of peo-
ple living on less than $1 a day dropped dramatically from
634 million in 1981 to 212 million in 2001.
The impact of economic growth is most evident in urban
areas. China’s cities are not only doing better than its rural
areas but are largely responsible for the country’s economic
boom, the effects of which are concentrated in the larger
cities. In 2001, per capita disposable income for urban resi-
dents was $829 compared to $278 for rural residents. In
1987, the income of the average urban household was
almost twice that of the average rural household; today it is
almost three times higher.
Sources: Linch & Zhi 2003; UN-HABITAT 2005a; TIME 2005; United Nations 2005a; Economy 2005; Worldwatch Institute 2006b; WHO/UNICEF 2000.
16
tions in the world – Shanghai, Beijing, Tianjin and Hong
Kong – since the mid-1980s, the country has been pursuing
an aggressive urbanization policy as a means of stimulating
both rural and urban economic development. The policy
aims to absorb the hundreds of millions of farmers who are
flocking to cities as a result of economic reforms and easing
of previously strictly enforced “urban residency permits”.
The residency control system is likely to be completely
eradicated in coming years as capital investments are made
to improve the infrastructure and economy of urban areas.
The aim of China’s pro-urban policies is to focus on the
development of towns and secondary cities to ease conges-
tion in the larger cities.
Equity Grants
Economic growth has also led to growing urban dispari-
ties. Prior to the economic reforms, the system made it dif-
ficult for villagers to migrate to cities, with the result that
slum formation was controlled, whenever possible. But
economic reforms saw a significant increase in migration
of unemployed workers and farmers to cities, with the
result that some inner-city and peri-urban areas have
been suffering from a gradual deterioration of living con-
ditions. In 2000, for instance, an estimated one-third of the
urban population in the country lacked adequate sanita-
tion. While the economic boom experienced by Chinese
cities induced investment in high and middle segments of
the housing market, it posed problems of affordability and
accessibility for families with limited income and savings.
Until the early 1980s, China’s urban housing market was
Some coastal cities, such as Shanghai, have skewed these
almost entirely the purview of state-owned enterprises
figures even further. In 2001, China’s largest city with a pop-
that were responsible for investing in and allocating hous-
ulation of 12.7 million had a gross domestic product (GDP)
ing within a strict command-and-control economy. High
of $4,510 per capita, almost five times the national average.
rates of urbanization and economic growth in the last two
The opening up of the Chinese economy has made
decades led to major macroeconomic reforms geared
Shanghai China’s most modern city and a favourite for for-
towards a “socialist economy based on market principles”
eign investment: in the 1990s, foreign investment in the city
and to the liberalization of the urban housing market in
totalled $45.6 billion. The city, which had only one sky-
the late 1990s.
scraper in 1988, today has more than 300. A mass transit sys-
tem, first-rate sea and river ports, well-developed railway
To facilitate low-income people’s access to the housing
and road networks and two international airports have
market, Chinese cities have been practising a policy of
increased the investment potential of the city and made it a
stimulating supply and demand through the use of equity
leading centre of international commerce and finance.
grants for people living in sub-standard housing. While
land remains the property of the state, leases are auc-
The prosperity of China’s cities is largely a result of econom-
tioned to developers to supply housing on a home owner-
ic reform policies that have a pro-urban focus. Although
ship basis. Low-income families living in slums or sub-
China already hosts 4 of the 30 largest urban agglomera-
standard housing are thus provided with once-in-a-life-
time equity grants based on the market value of their
existing housing, which enables them to access mortgage
instruments. Developers, on the other hand, are provided
incentives in the form of tax reductions or exemptions.
The use of equity grants, combined with incentives for
housing developers to provide affordable housing, led to
the production of more than 20 million housing units in
the last five years. Chinese cities are hoping to avert the
proliferation of urban ghettos and slums by providing
more affordable housing. In large housing estate develop-
ments, many of which attract foreign direct investment, a
new level of self-governance has also emerged, with resi-
dents electing committees to oversee and manage urban
safety and security, environmental conservation and the
needs of youth and the elderly.
Prosperity and Pollution
China’s recent gains in economic growth and industrial-
ization have in many cases exacerbated environmental
problems in its cities. Economic growth has increased
consumer purchasing power, with the result that Chinese
cities, such as Beijing – once the bicycle capital of the
world – are now teeming with motor vehicles, a leading
cause of air pollution. There are 1.3 million private cars in
Beijing alone, an increase of 140 per cent since 1997.
Experts believe that China’s skyrocketing private car own-
ership and lax implementation of industrial emission reg-
ulations could threaten the recent gains it has achieved
on the economic front. China’s manufacturing-based
economy has made it one of the world’s largest con-
sumers: in 2005, the country used 26 per cent of the
world’s crude steel, 32 per cent of its rice and 47 per cent
of its cement.
According to the World Bank, China is home to 16 of the 20
most polluted cities on the planet. China is also the second
largest producer of greenhouse gases, after the United
States. Environmental degradation robs the nation of up
to 12 per cent of its GDP, and every year some 400,000
Chinese die prematurely of respiratory illnesses and some
30,000 children die from diarrhoea caused by drinking
unclean water. Towns and villages along China’s most pol-
luted rivers are also reporting more cases of cancer and
miscarriages. According to the Yellow River Conservancy
Commission, river pollution costs the country $.1.9 million
annually. If China is to sustain its remarkable economic
growth, it must also ensure that its cities are sustainable.
17
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

1.2 Putting Slums on the Map: A Global and Regional Overview

Global trends

I
n the last decade, an increasing number of governments
around the world have enlarged democratic space with-
in their countries and opened up their markets in
response to the demands of a globalizing world. But
democracy and market economies have not had the
desired effect of reducing inequalities within and among the
world’s regions. On the contrary, from 1960 to 1999, the
incomes of the richest countries grew to exceed those of the
poorest by 35 times.1 Economic growth, it turns out, does not
automatically result in prosperity for all. In many countries,
national gross domestic product (GDP) rates have risen much
more quickly than national poverty rates have fallen; the effect
has been a growing gap between the rich and the poor, a gap
that is most evident in cities of the developing world.
Inequality has a direct bearing on patterns of urbanization.
The rich in most countries live a world apart from the poor, with
homes in protected urban enclaves and access to the latest tech-
nology, the best services and the most comfort. The rest, especial-
ly slum dwellers, live in the most deprived neighbourhoods,
struggling to gain access to adequate shelter and basic services,
such as water and sanitation. Many slum dwellers also live under
the constant threat of eviction. Such stark differences and divi-
sions can be found among regions and countries, but also with-
in countries and cities. Especially in the developing world, urban
zones of poverty and despair commonly skirt modern cosmopol-
itan zones of plenty. If current trends are not reversed, cities will
become more and more spatially divided, with high- and middle-
income residents living in the better-serviced parts of the city,

TABLE 1.2.1 POPULATION OF SLUM AREAS AT MID-YEAR, BY REGION; 1990, 2001, 2005 AND ANNUAL SLUM GROWTH RATE

% Slum % Slum % Slum Slum

slum Population slum Population slum Population annual
(thousand) (thousand) (thousand) growth rate
1990 1990 2001 2001 2005 2005 (%)

Region

WORLD 31.3 714,972 31.2 912,918 31.2 997,767 2.22

Developed regions 6.0 41,750 6.0 45,191 6.0 46,511 0.72

EURASIA (Countries in CIS) 10.3 18,929 10.3 18,714 10.3 18,637 -0.10
European countries in CIS 6.0 9,208 6.0 8,878 6.0 8,761 -0.33
Asian countries in CIS 30.3 9,721 29.4 9,836 29.0 9,879 0.11
Slum overlooking Mumbai, India R. A. ACHARYA/UNEP/STILL PICTURES

Developing regions 46.5 654,294 42.7 849,013 41.4 933,376 2.37

Northern Africa 37.7 21,719 28.2 21,355 25.4 21,224 -0.15
Sub-Saharan Africa 72.3 100,973 71.9 166,208 71.8 199,231 4.53
Latin America and the Caribbean 35.4 110,837 31.9 127,566 30.8 134,257 1.28
Eastern Asia 41.1 150,761 36.4 193,824 34.8 212,368 2.28
Eastern Asia excluding China 25.3 12,831 25.4 15,568 25.4 16,702 1.76
Southern Asia 63.7 198,663 59.0 253,122 57.4 276,432 2.20
South-Eastern Asia 36.8 48,986 28.0 56,781 25.3 59,913 1.34
Western Asia 26.4 22,006 25.7 29,658 25.5 33,057 2.71
Oceania 24.5 350 24.1 499 24.0 568 3.24

Source: UN-HABITAT 2005, Global Urban Observatory, Urban Indicators Programme, Phase III.

18 19
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

What is a slum?
At an Expert Group Meeting in
November 2002, UN-HABITAT and its
partners came up with a provisional
definition of “slum”: a settlement in
an urban area in which more than
half of the inhabitants live in
inadequate housing and lack basic
services. Developing an operational
definition – one with measurable
indicators – required further
refinement, recognizing that slums
can be geographically contiguous or
isolated units. UN-HABITAT
Low-income housing in Mexico City, Mexico RON GILING/STILL PICTURES
A slum household is a group of individuals living
under the same roof in an urban area who lack
one or more of the following five conditions:
Durable housing: A house is considered “durable” if
it is built on a non-hazardous location and has a struc-
ture permanent and adequate enough to protect its
inhabitants from the extremes of climatic conditions,
such as rain, heat, cold and humidity.
Sufficient living area: A house is considered to
provide a sufficient living area for the household
members if not more than three people share the
same room.
Access to improved water: A household is consid-
Access to sanitation: A household is considered to
have adequate access to sanitation if an excreta dis-
posal system, either in the form of a private toilet or
a public toilet shared with a reasonable number of
people, is available to household members.
Secure tenure: Secure tenure is the right of all
individuals and groups to effective protection
against forced evictions. People have secure tenure
when there is evidence of documentation that can
be used as proof of secure tenure status or when
there is either de facto or perceived protection
against forced evictions.

therefore focuses on the household
as the basic unit of analysis. A
single operational definition of slums
is used throughout this Report.*
ered to have access to improved water supply if it
has a sufficient amount of water for family use, at an
affordable price, available to household members
without being subject to extreme effort, especially
on the part of women and children.
* This definition may be amended according to the situa-
tion in a specific city. For example, in Rio de Janeiro, living
area is insufficient for both the middle classes and the
slum population and is not a good indicator. It could either
be omitted, or it could be combined with another indicator
to denote two or more shelter deprivations.

TABLE 1.2.2 URBAN AND SLUM GROWTH RATES BY REGION
Urban growth significantly
higher than slum growth
The findings presented in the following chapters represent a ■ Slums: The emerging human settlements of the
new approach to measuring and understanding slums, developed 21st century?
by UN-HABITAT in response to the international community’s
recognition in 2000 that slums cannot be considered an unfortu- The word “slum” first appeared in 19th century London,
nate by-product of urbanization, but instead need to be addressed when the burgeoning urban working classes moved into over-
comprehensively as a major development issue. Approaching crowded and poorly serviced tenements, living close to the fac-
slums as a specific type of human settlement with discernable tories and industrial plants that employed them. The term
characteristics and impacts on the people who live in them pro- referred to what was initially designated “a room of low repute”,
vides a framework for moving toward Millennium Development but over time took on the generic definition, “a squalid and
Goal 7, target 11: by 2020, to have achieved a significant improve- overcrowded urban area inhabited by very poor people”.2
In 2005, there were 998 million slum ment in the lives of at least 100 million slum dwellers Although slums continued to grow over the course of the last
Urban Slum dwellers in the world; if current
growth two centuries, their evolution was particularly swift in the latter
growth
Regions rate rate trends continue, the slum population half of the 20th century as the developing world became more
will reach 1.4 billion by 2020. FIGURE 1.2.1 SLUM POPULATIONS, 1990-2020 urbanized. Today’s slums are much larger and have many more
residents than the slums prevalent in 19th century Europe and
North America. The slum population of Rio de Janeiro, for
5,000,000 example, is almost the same size as the total population of
Latin America and 2.21 1.28 Helsinki. Mumbai’s more than 5 million slum dwellers exceed
the Caribbean and the poor living in spatially or socially segregated slums the total population of Nairobi. Slum dwellers now live prima-
4,000,000
Northern Africa 2.48 -0.15 with few services or none at all. rily in the cities of Africa, Asia and Latin America, although a
Eastern Asia 3.39 2.28 Not all of the world’s urban poor live in slums; poverty in smaller number also live in cities of the developed world.
South-Eastern Asia 3.82 1.34
Population ('000)

cities has various social and economic dimensions that have 3,000,000 Slum housing ranges from crowded tenement buildings in
little to do with the physical structure of the houses or the Hong Kong to mud-and-tin shacks in Cape Town. Slums can
Urban and slum environments in which people live. Conversely, not all be inner-city tenements in cities of the developed world, shan-
2,000,000
growth similar Western Asia 2.96 2.71 those who live in slums are poor – many people who have ty towns on the periphery of large cities or densely packed
Southern Asia 2.89 2.20 risen out of income poverty choose to continue living in neighbourhoods bordering high-income areas. Individual
Sub-Saharan Africa 4.58 4.53 1,000,000
slums for various reasons ranging from lack of affordable households located in high- or middle-income neighbour-
housing in better parts of the city to proximity to family hoods may also fit the definition of slums. In some parts of the
Developed world 0.75 0.72 and social networks. However, if the quality of housing and 0 developing world, gradations of slums are common, with each
the existence of basic services are used as criteria to deter- 1990 2001 2005 2010 2015 2020 variation having a different name. For instance, in India, a
World 2.24 2.22 mine poverty levels, then slums represent a physical dimen- Year
chawl (a densely packed block of one-room “apartments” with
World Developing Developed & Eurasia
sion of poverty. This aspect of urban poverty is the focus of shared toilets and bathrooms) is quite different from a zopad-
Source: UN-HABITAT 2005, Global Urban Observatory. this Report. Source: UN-HABITAT, Global Urban Observatory 2005. patti (a shack made of non-durable materials, often located in

20 21
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
a crowded slum settlement within or on the outskirts of a
city). Yet, both types of housing could fall under UN-
HABITAT’s definition of slum households if they lack one or
more of the five conditions that are necessary to deem a house
“adequate” (see box).
One out of every three city dwellers – nearly one billion peo-
ple – lives in a slum today. For many years, governments and
local authorities viewed slums as transient settlements that
would disappear as cities developed and as the incomes of slum
dwellers improved. However, evidence shows that slums are
growing and becoming permanent features of urban land-
scapes. Slums have carved their way into the fabric of modern-
day cities, making their mark as a distinct category of human
settlement that constitutes a space between “rural” and
“urban”. Given the proliferation of slums around the world
and the growth of city-regions, in which larger cities act as
centres for smaller cities and towns and a large rural hinter-
land, the simple duality that exists in the traditional framing
of human settlement patterns no longer suffices to describe the
reality of people’s lives. Slums in many cities are no longer just
marginalized neighbourhoods housing a relatively small pro-
portion of the urban population: in some cities, particularly in
Southern Asia and sub-Saharan Africa, slums host significant-
ly large proportions of the urban population and slum growth
is virtually synonymous with urbanization; this calls for new
ways of looking at cities and the slums within them.
Will slums become a predominant type of settlement in the
21st century? If no preventive or remedial action is taken, they
may indeed come to characterize cities in many parts of the
developing world.
The vast majority of slums, more than 90 per cent, are locat-
ed in cities of the developing world, where urbanization has
become virtually synonymous with slum formation. This is
especially so in sub-Saharan Africa, Southern Asia and Western
Asia, where urban growth over the last 15 years has been
accompanied by a commensurate growth in slums.
Slum and urban growth rates are highest in sub-Saharan
Africa, 4.53 per cent and 4.58 per cent per year, respectively –
nearly twice those of Southern Asia, where slum and urban
growth rates are 2.2 per cent and 2.89 per cent per year, respec-
tively. In Western Asia, annual slum and urban growth rates are
quite similar, at 2.71 per cent and 2.96 per cent respectively,
while in Eastern Asia and Latin America, slum growth rates are
significantly lower than urban growth rates, although slum
growth rates are relatively high in both regions: 2.28 per cent
and 1.28 per cent per year, respectively. South-Eastern Asia and
Northern Africa are two regions where slum growth has not
kept pace with urbanization; in both regions the proportion of
slum dwellers has actually declined in recent years from over 36
per cent of the urban population in 1990 to approximately 25
per cent in 2005. Eastern Asia and Latin America and the
Caribbean also have urban growth rates that are higher than
slum growth rates. This suggests that countries within these
regions have in recent years taken active steps to reduce the
number of slum dwellers or prevent slum formation.
22
Kibera, Nairobi HIROSHI SATO
At the global level, 31.2 per cent of all urban dwellers lived
in slums in 2005, a proportion that has not changed signifi-
cantly since 1990. However, in the last 15 years, the magni-
tude of the problem has increased substantially: 283 million
more slum dwellers have joined the global urban population.
In 1990, there were nearly 715 million slum dwellers in the
world.3 By 2000 – when world leaders set the target of
improving the lives of at least 100 million slum dwellers by
2020 – the slum population had increased to 912 million. In
2005, there were almost 1 billion (998 million) slum dwellers
in the world; if current trends continue, UN-HABITAT esti-
mates that the slum population will reach 1.4 billion by 2020.
Trends in developing regions
■ Slum trends in Africa
An interesting disparity exists between Northern Africa and
sub-Saharan Africa in terms of slum growth and slum preva-
lence: while the former is experiencing negative slum growth,
the latter is experiencing the opposite trend, with extremely
high slum growth rates of 4.53 per cent per year – the highest
in the world.4
Northern Africa achieved a reduction in both the number
and proportion of slum dwellers between 1990 and 2005. The
share of slum dwellers in the region fell from 37.7 per cent to
25.4 per cent, with the absolute number of people living in
slums decreasing by half a million, to just over 21 million. The
reduction may be attributed to the relatively low levels of slum
prevalence in the region in general, as well as to the implemen-
tation of policies aimed at reducing the number of slum
dwellers within countries. Countries such as Morocco and
Tunisia have been very successful in improving the lives of
slum dwellers, while Egypt, where slums exist on a much larg-
er scale, was able to address the problem with pro-poor poli-
cies and substantial investments in improving the shelter con-
ditions of people living in cities.
Sub-Saharan Africa, on the other hand, has been unable to
manage or reduce slum growth. This is partly attributed to the
declining economies of some countries in the region, coupled
with its disproportionate share of HIV prevalence and con-
flicts, both of which have exacerbated slum formation and
worsened living conditions in cities. Slums in cities such as
Khartoum, for instance, have grown remarkably in the last
decade, largely due to an influx of internally displaced persons
(IDPs) from Southern Sudan, which has been suffering from a
protracted civil war.
Sub-Saharan Africa has the highest proportion of slum
dwellers in the world: 71.8 per cent. In terms of absolute num-
bers, it is home to the third most populous slum population
among the regions of the developing world, after Southern
Asia and Eastern Asia. In the last 15 years, the number of slum
dwellers in sub-Saharan Africa has almost doubled, from 101
MAP 1 URBAN POPULATION AND SLUM PROPORTION IN AFRICAN COUNTRIES, 2001
Urban Population
<= 4 Mio
>4-10 Mio
>10-30 Mio
>30-75 Mio
>75-150 Mio
>150 Mio
Percentage of
Urban Slum Population
<=10%
>10-30%
>30-60%
>60%
Source: UN-HABITAT, Global Urban Observatory 2005.
million in 1990 to 199 million in 2005. Urban growth in the
region is almost identical to slum growth – a trend that is also
prevalent to a lesser extent in other regions. Given the high
slum growth rate in the region, the number of slum dwellers
will likely double by 2020, reaching nearly 400 million, and
overtake the slum populations of both Southern Asia and
Eastern Asia, which are estimated to rise to 385 million and
299 million, respectively.
In many of the region’s countries, notably Angola, Ethiopia,
Mali, Mauritania, Sudan, and Tanzania, slum populations are
expected to double within the next 15 years. Slum households
are the norm rather than the exception in many cities. For
instance, UN-HABITAT’s Urban Inequities Survey,5 con-
ducted in the capital of Ethiopia, Addis Ababa, has shown
that less than 10 per cent of that city’s inhabitants live in non-
slum areas.
When slums constitute the largest proportion of a city, dif-
ferentials between, even within, slums also become apparent.6
As in Ethiopia, several other primarily rural sub-Saharan
African countries in the first stages of their urban transition –
including Chad, and the Central African Republic – have very
high proportions of slums in their cities. Not surprisingly, the
living conditions in slums within these countries are also
23
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

extremely severe, as 70 per cent to 90 per cent of households Ninety per cent – or 195.7 million people – of Eastern impact on both the magnitude and the proportion of slums.
are deprived of more than two basic shelter needs, such as Asia’s slum dwellers live in China. Chinese slum dwellers The main reasons behind Thailand’s ability to reduce slum
water or sufficient living space. This trend applies to much of account for 20 per cent of the world’s total. It is important to growth are a strong political commitment by its leadership,
sub-Saharan Africa, where slum households are likely to lack note that since 1990, China has been held up as a success in accompanied by a tradition of strategic planning and monitor- Most of the slum dwellers in Southern
water, sanitation, durable housing, sufficient living space, and increasing the scale of low-cost housing schemes, thus prevent- ing development efforts, which have been an integral part of Asia – 63 per cent, or almost 170 million
secure tenure altogether, or a combination of at least three of ing slums before they even form.8 Despite such measures, the the development tradition for the last 30 years. Many of the people – reside in India.
these indicators of adequate housing. country suffers from high levels of slum prevalence. There sub-region’s countries also have an active civil society.
could be two reasons for this, other than the fact that the total Although the high slum growth rates in Cambodia and Lao
population of China constitutes one fifth of the world’s popu- People’s Democratic Republic suggest a less optimistic future,
■ Slum trends in Asia lation. One is the need for a lapse in time for slum prevention there is evidence that they might be able to curb slum growth,
policies to have an impact on the ground. The other could be as they have recently initiated slum prevention policies.
In absolute numbers, Asia has the largest share of the world’s the mismatch between UN-HABITAT and national defini- Western Asia, on the other hand, lags far behind the other an increase in its slum population. Both countries have relative-
slum population – in 2005, the region was home to more than tions of what constitutes a slum. While UN-HABITAT con- sub-regions in terms of slum prevention. Slum and urban ly small populations, so the reduction or increase in slums there
half the world’s total slum population, or about 581 million siders the de facto status of dwellings in the cities of China, growth rates in the sub-region are almost the same, reaching might not make a dent in the overall slum figures by 2020, but
people. Some sub-regions within Asia are faring worse than irrespective of their legal status, national authorities do not nearly 3 per cent per year. The countries of Western Asia have progress is still important, as it would indicate greater stability
others. Eastern and Southern Asia harbour 80 per cent of the consider people who live outside the de jure residential area or made little progress on any of the Millennium Development in the sub-region, accompanied by better social indicators.
slum dwellers in the region, with Southern Asia hosting near- those who do not possess residency permits as bona fide resi- Goal indicators and have not been able to sustain the momen- Slum growth in the largest country of the sub-region, Turkey,
ly half the region’s slum population. These figures are largely dents of a city.9 tum of development they gained between 1980 and 1990, as declined radically between 1990 and 2001, from 23.3 per cent
attributable to China and India, which are the most populous It is interesting to note that real success stories in the region, the region has in recent years been engulfed in political turmoil to 17.9 per cent, primarily because of an effective policy of
countries in the world and have significant proportions of in terms of decreasing slum growth significantly, have occurred that has exacerbated the refugee crisis and worsened conditions decentralization, which empowered the municipal govern-
their urban populations living in slum conditions. Although in South-Eastern Asia, in countries such as Thailand, where in cities. In countries such as Jordan, slums have grown at the ments to borrow directly from international financial institu-
China hosts the world’s largest slum population – almost 196 policies implemented even before the 1990s have had a strong rate of 4.3 per cent per year, and Lebanon has also experienced tions to build or upgrade water and sanitation networks.
million people – its slum prevalence in 2001 was lower than
that of India; UN-HABITAT estimates that 38 per cent of
China’s urban residents lived in slum conditions that year,
MAP 2 URBAN POPULATION AND SLUM PROPORTION IN ASIAN COUNTRIES, 2001
compared with India’s 56 per cent.
Most of the slum dwellers in Southern Asia – 63 per cent,
or almost 170 million people – reside in India. The share of
Southern Asia’s slum dwellers constitutes 27 per cent of the
global total; India alone accounts for 17 per cent of the world’s
slum dwellers. India has pioneered many best practices and
good policies in recent years that are having some impact on
the lives of slum dwellers, but they have not reached a suffi- Mumbai RASNA WARAH
cient scale to ameliorate the proliferation of slums. Although
the country has seen remarkable economic growth rates in
recent years and has managed to reduce extreme poverty by 10
per cent in the last decade, the impact of poverty reduction is
still not being felt in cities. Unless more radical policies are
pursued in India, the global target for improving the lives of
slum dwellers will not be reached.
Other countries that need to address this challenge urgently are
Bangladesh and Pakistan which, along with India, have among
the highest urban poverty rates and the largest urban populations
in the sub-region. UN-HABITAT data shows that Bangladesh
was home to 30 million slum dwellers in 2001, and 85 per cent
of its urban population lived in poverty that year; 74 per cent of
Pakistan’s urban population lived in poverty in 2001 – more than Urban Population Percentage of
Urban Slum Population
35 million people. Through successful initiatives such as the <= 4 Mio
<=10%
Orangi project,7 Pakistan has demonstrated how the lives of slum >4-10 Mio
dwellers can be improved at the local level. Three decades on, >10-30 Mio >10-30%
however, the project has not been able to scale up its interventions >30-75 Mio >30-60%
to have a national impact. Bangladesh’s development campaigns >75-150 Mio
>150-472 Mio >60%
through the Bangladesh Rural Advancement Corporation and its
Grameen Bank initiative have focused on alleviating rural pover-
ty, so have had negligible impact in urban areas. Source: UN-HABITAT, Global Urban Observatory 2005.

24 25
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Mumbai’s Quest for ‘World City’ Status
Mumbai, the capital city of the state of Maharashtra and
India’s most important financial capital, has a population of
18.3 million people, making it the fourth largest urban
agglomeration in the world, after Tokyo, Mexico City and New
York-Newark. The city hosts one of the world’s largest slum
populations: more than 5 million of the city’s residents are
slum dwellers. More people live in Mumbai’s slums than in the
entire country of Norway.
Despite its large slum population – or, as some would argue,
because of it – Mumbai has emerged as one of India’s leading
commercial and cultural centres, home to the country’s huge-
ly successful film industry and a booming stock exchange.The
city alone pays almost 40 per cent of the nation’s taxes.
There is a perception among many of the country’s policy-
makers, however, that Mumbai’s progress is being hampered
by its image as a city of slum dwellers, which severely erodes
its ambitions to become the “Shanghai of India” with clean
streets, gleaming skyscrapers and a modern and efficient
transport and communications network. This perception was
reinforced in late 2004 when the Indian government
embarked on a campaign to make Mumbai a “world-class
city” in response to a call by the Indian Prime Minister in
October 2004 to transform Mumbai into an international hub
for trade and commerce. The call was supported by
Maharashtra’s chief minister, who submitted an ambitious
four-year, $8 billion proposal for modernizing Mumbai, which
included the building of new roads, a subway system and a
large-scale public housing project. The modernization pro-
posal followed an earlier Slum Rehabilitation Scheme in the
mid-1990s that aimed to improve the lives of 4 million slum
dwellers through public-private partnerships that involved
builders in the private sector and the authorities.
Demolition drive
In late 2004, despite progressive slum improvement and
tenure regularization policies and programmes, the govern-
ment of Maharashtra began a slum demolition drive aimed
at removing slums and shanty towns in the city. Between
December 2004 and March 2005, more than 90,000 shanties
were torn down, in violation of poll promises, international
covenants to which India is a signatory and a 2001 Slum
Areas Act, which protected all slums built prior to 1995. (The
Act stipulates that all slum dwellers who could establish that
Sources: Moreau and Mazumdar 2005; Risbud, 2003; YUVA and Montgomery Watson Consultants 2001; Sharma 2005a and 2005b; Burra 2005; Arputham 2001;
Mehta 2004; Kothari 2005; United Nations Department of Economic and Social Affairs, Population Division 2004.
26
their names were on the electoral roll on 1 January 1995
were protected, to the extent that their homes could not be
demolished without rehabilitation.)
Amid public outcry and pressure from the ruling Congress
party, the demolitions were halted in February 2005, but
many believe that the plan to make Mumbai a world-class
city is still very much on the cards. “The reality of course is
that a new Mumbai cannot be built on the corpses of its
poor, the very people who hold up this city,” argued journal-
ist Kalpana Sharma of The Hindu newspaper.
Jockin Arputham, founder of India’s National Slum Dwellers
Federation, has consistently argued that it is Mumbai’s poor,
who allow the city to flourish by providing cheap labour and
services. “The poor work as refuse collectors, construction
labourers, handcart pullers, vegetable vendors, factory work-
ers, domestic workers and so on.They provide goods and serv-
ices at rates that most of the city’s people can afford. But when
it comes to their housing, the city turns its back on them.”
Dire slum conditions
Slum children, Mumbai MARK EDWARDS/STILL PICTURES
Despite the active role played by non-governmental organiza-
tions (NGOs) and slum federations in the city, the situation of
slum dwellers in Mumbai remains dire. A recent survey for the
Mumbai Sewerage Disposal Project found that 42 per cent of
slum dwellings in the city had an area of less than 10 square
metres and only 9 per cent had an area of more than 20 square
metres. Almost half of the households in slums got their water
from shared standpipes and only 5 per cent had direct access
to water through individual taps.The city’s sanitation situation
was even more alarming: 73 per cent of the city’s slum house-
holds – housing 3.86 million residents – depended exclusively
on public toilets. Moreover, overuse and poor maintenance and future generations of slum dwellers and migrants. People are still starving to death in other parts of India. In
had made public toilets a health hazard, especially in areas (Currently, only 58,000 new low-cost housing units are avail- Bombay, there are several hundred slimming clinics.”
where the user group was undefined.Less than one per cent of able for pre-1995 slum dwellers.) The report provides the (Bombay was officially renamed Mumbai in 1996, but like
the slum population had access to individual toilets or to pay- framework for the city’s urban renewal scheme,which,if imple- many diehard Bombayites, Mehta prefers to call it by its
per-use toilets constructed by private agencies or NGOs. mented, will cost upwards of $40 billion over the next decade. old name.)
City authorities are already looking into how the funds can
be raised from federal and local governments and from Others are cynical about Mumbai’s attempts to reach
Despite the daunting conditions in its slums,Mumbai is a mag-
international lending institutions. world-class status. “One hardly needs to emphasize that
net for Indians, not only from neighbouring cities and villages,
the world over, cities with good and affordable public
but also from the rest of the country. According to “Vision
Author Suketu Mehta feels that Mumbai cannot escape transport are also the most liveable … . The best cities in
Mumbai”, a 2003 report by the private consultancy firm
the demands of globalization, including the pressures to the world are also the ones that have affordable housing
McKinsey & Co, the city urgently needs to build at least 1.1 mil-
become “world class”. In many ways, he writes, the city has for all classes,” writes Sharma, who is the author of a book
lion affordable housing units in the next decade for current
already parted with the rest of India as the gap between on Dharavi, Mumbai’s largest slum. “We need to put aside
the haves and the have-nots widens: “In the Bayview Bar of our obsession with becoming ‘world class’. Let us make our
the Oberoi Hotel you can order a bottle of Dom Pérignon cities liveable for all the people. That itself is a big agenda
for one and a half times the average annual income … . for the future.”
27
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

Slums: The shelter dimension of urban poverty Defining and Monitoring Slums: Seeing Beyond the Stereotypes
Current debates over the multidimensional aspects poverty. These include: human assets (such as skills poverty, with some experts claiming that it has been Until recently, empirical evidence regarding liv- Shelter deprivation indicators Using the first four slum definition indicators,
of poverty recognize that income-based poverty and good health), natural assets (such as land), grossly underestimated. For instance, official fig- ing conditions in the world’s slums was not it has been possible to estimate the preva-
measurements do not capture the scale or range of physical assets (such as access to roads and other ures show that only 9.7 per cent (or 14.6 per cent, available in a universally comparable format. Slum dwellers often live in difficult social and lence and magnitude of slums and to calcu-
poor living conditions experienced by people around infrastructure), financial assets (such as access to depending on the survey used) of Phnom Penh’s Slums were the “invisible” parts of cities – nei- economic conditions that manifest different late projections in most countries of the world
the world. The link between income and levels of savings and credit) and social assets (such as net- population lived below the poverty line in 1999, yet ther reflected in official data or maps, nor recog- forms of deprivation – physical, social, eco- using existing household surveys and census-
deprivation is weak and misleading, as many who works of family and other contacts that can be 2001 data shows that an estimated 40 per cent of nized by authorities. Speculative analysis sug- nomic and political. Four out of five of the es, including Demographic and Health Surveys
live above the poverty line may suffer from serious called upon in times of need). Capturing the depth the Cambodian capital’s population lived in informal gested that people living in slums were experi- slum definition indicators measure physical and UNICEF Multiple Indicator Cluster
deprivations in other areas, while those below the and magnitude of urban poverty is particularly settlements, or slums. Similar gaps between official encing a continuous deterioration of their living expressions of slum conditions: lack of water, Surveys, conducted between 1990 and 2001.
poverty line may suffer from income poverty, but important when monitoring the achievement of figures for urban poverty levels and slum estimates environments, yet figures were often inaccurate lack of sanitation, overcrowding, and non- These indicators are considered “shelter dep-
may not be “poor” in other aspects. Millennium Development Goal 7, target 11. in other cities suggest that poverty in cities is still or contradictory within and among countries. durable housing structures. These indicators rivation indicators” by UN-HABITAT and its
being viewed through the income lens that does not Sometimes, figures overestimated or underesti- focus attention on the circumstances that sur- partners.
These debates are highly applicable to measure- Urban poverty and slums take into account the living conditions of city mated the reality of the situation, based on sub- round slum life, depicting deficiencies and
ments of urban poverty. UN-HABITAT and others dwellers and their needs for goods and services jective concepts about the nature of slums or the casting poverty as an attribute of the environ- The physical and visible manifestations of
have consistently argued that understanding the In this Report, UN-HABITAT attempts to analyse a that are specific to urban areas. While not all peo- motivation of researchers or political entities ments in which slum dwellers live. housing that lacks basic services, space and
various dimensions and degrees of urban poverty is particular dimension of urban poverty that has not ple who live in slums necessarily live below the involved in estimation. The production of data security take many forms, resulting in diverse
important in order to construct pro-poor policies been adequately captured in either national statis- poverty line, the huge discrepancy between these was thus not the result of a reliable monitoring The fifth indicator – security of tenure – has to types of slums. Not all are as easily distin-
that have a tangible impact on the living conditions tics or in United Nations data – that of shelter depri- two figures suggests that the poverty line used in process. do with legality, which is not as easy to meas- guishable or visible as the shanty towns
of the urban poor. UN-HABITAT is convinced that vation. This shelter dimension of urban poverty is cities is not realistic. ure or monitor, as the status of slum dwellers cramped together on the periphery of cities
neither the food basket nor the one-dollar-a-day measured using five key indicators: access to water; Global monitoring has also faced a political often depends on de facto or de jure rights – such as Mumbai, Nairobi or Cape Town. In
indicator can accurately reflect the diverse experi- access to sanitation; durability of housing; sufficient Until a universal knowledge base is developed to obstacle over the term “slum”, which has often or lack of them. This indicator has special rele- some places, slums are less visible to the eye:
ences of people living in poverty in both rural and living area; and secure tenure. produce information on other forms of urban pover- been deemed derogatory by urban planners, city vance for measuring the denial and violation dwellings may look durable or permanent
urban areas. ty, the thrust of UN-HABITAT’s discussion remains authorities and slum dwellers themselves. The of housing rights, as well as the progressive from the outside, but living conditions inside
In cities, poverty is quite often physically and spa- focused on the knowledge and data it has gathered preferred terms, including “informal settlement”, fulfillment of these rights. There currently the dwelling may portray another picture. For
The problem with current measurements of urban tially visible in slums, which suffer from poor quality, and generated on shelter deprivations, which tend “squatter settlement” and “unplanned neigh- exists no mechanism to monitor secure tenure instance, many multi-storey public housing
poverty, and consequent policy discussions, is the insecure, hazardous and overcrowded housing and to be concentrated in slums. This is being done bourhood”, have been used interchangeably with as part of Millennium Development Goal 7, projects at the periphery of urban cores or old,
division of urban populations into the “poor” and the lack infrastructure and basic services. While shelter bearing in mind that “slum dwellers” neither repre- “slum”, but have not heretofore been linked to target 11, as household-level data on property dilapidated buildings in inner cities could
“non-poor” with little recognition of the diversity deprivations are most apparent in slums, they do sent all individuals who live in poverty in cities, nor specific indicators regularly reported on by gov- entitlement, evictions, ownership, and other qualify as slums if they have been neglected
within the “poor” and the “almost poor” with regard not necessarily denote levels of urban poverty and are all slum dwellers, as a heterogeneous group, ernments and stakeholder organizations. The dif- indicators of secure tenure is not uniformly or ill-serviced for significant periods of time,
to their deprivations, vulnerabilities and needs. This are only a subset of a wider range of urban poverty “living below the poverty line”. Nonetheless, slums ficulty in developing a workable measurement available through mainstream systems of data as would many workers’ hostels or dormito-
measurement also fails to recognize that people experiences. However, the huge gap between offi- are a good starting point – but not the ultimate or strategy and the lack of reporting on slums illus- collection, such as censuses and household ries. Such places typically do not look like
have a variety of assets, which may or may not cial urban poverty figures and the proportion of peo- definitive point – for describing urban poverty and trates that they are conceptually complex and surveys. In this Report, UN-HABITAT points to slums, but if their residents experience some
translate into income or cash, but which nonethe- ple living in slums does raise questions about the capturing the scale and depth of shelter depriva- methodologically elusive. trends that suggest levels and severity of inse- form of shelter deprivation or insecure tenure,
less play an important role in determining levels of validity of the methodologies used to measure urban tions in cities. cure tenure around the world, but these fig- then, according to the UN-HABITAT definition
Different cultures and countries define the physi- ures are based on secondary sources that may of a slum household, these residents qualify
Sources: World Bank 2000/2001; Jonsson & Satterthwaite 2000, Asian Coalition for Housing Rights 2001. cal and social attributes of slums differently. UN- or may not reflect the reality on the ground. as slum dwellers.
HABITAT acknowledges this diversity and the
fact that slums take many different forms and
names. Bearing this in mind, in 2002, UN-HABI-
Various dimensions of urban poverty TAT, the United Nations Statistical Division and Shelter deprivations Visible and non-visible types of slums
the joint UN-HABITAT/World Bank Cities and security of tenure
Inadequate and often unstable Inadequate, unstable or risky which can be easily accessed when Voicelessness and powerlessness
Alliance gathered together a group of experts to Slum stereotypes Atypical slum households
income, which impacts people’s asset base (non-material and basic necessities are no longer within non-responsive political
define slums and propose a way to measure
ability to pay for non-food items, material) including lack of assets affordable, such as public housing systems and bureaucratic Shanty towns Dilapidated Multi-family Multi-story Multi-story
them. The resulting definition and methodology at urban houses flats in public dormitories
such as transport, housing and that can help low-income groups and free medical services. structures, leading to little or no
represent a compromise between theoretical and periphery within cities buildings housing
school fees. cope with fluctuating prices or possibility of receiving entitlements to
methodological considerations. The agreed-upon (single units, (single units, within city projects
incomes, such as lack of access to Inadequate protection of rights goods and services; of organizing,
definition is simple, operational and pragmatic: one story) one story) centers
Poor quality, hazardous, land or credit facilities. through the operation of the law, making demands and getting a fair
it can be easily understood and adapted by gov-
overcrowded, and often insecure including regulations and response; and of receiving support for Lack of access to water x x
ernments and other partners; it offers clear,
housing Inadequate public procedures regarding civil and developing initiatives. Also, no means
measurable indicators, provided as a proxy to Lack of access to sanitation x x x x
infrastructure, such as schools political rights, occupational health of ensuring accountability from aid
capture some of the essential attributes of
Inadequate provision of basic and hospitals. and safety, pollution control, agencies, NGOs, public agencies and Non-durable housing x x x
slums; and it uses household-level data that is
services (piped water, sanitation, environmental health, protection private utilities, and of being able to
collected on a regular basis by governments, Insufficient living space x x x x
drainage, roads, footpaths, etc.) which Limited or no safety nets to from violence and forced evictions participate in the definition and
development agencies and non-governmental
increases the health burden and often ensure basic consumption can be and, protection from discrimination implementation of urban poverty Insecure tenure x x x
organizations, which is accessible and available
the work burden. maintained when incomes fall and and exploitation. programmes.
in most parts of the world.

Adapted from Satterthwaite 2004.

28 29
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

Woman-headed households in cities Age pyramids for slum and non-slum populations in Brazil and South Africa
Although “the proportion of woman-headed In Latin America, most women heading house-
households” is not among the indicators used to holds live in cities. With the exception of Haiti, a 80+
75-79
80+
75-79 males as more men migrate to cities than
monitor progress on Millennium Development majority of these households are located in non- 70-74 70-74
65-69
60-64
65-69
60-64 women, particularly in sub-Saharan Africa. Yet,
Goal 3 on promoting gender equality and empow- slum areas of cities. The overall share of urban
55-59
50-54
Male Female 55-59
50-54
Male Female
45-49 45-49 when the pyramids of these two countries are
ering women, there is a general belief that households headed by women in Haiti is quite
40-44
35-39
40-44
35-39
30-34 30-34 compared with the age group of the slum popu-
woman-headed households deserve special atten- 25-29 25-29
high (50 per cent), compared with 38 per cent in 20-24
15-19
20-24
15-19 lation, the findings are noteworthy: instead of
tion as they fall under the category of the poorest 10-14
5-9
10-14
5-9
rural areas. 0-4 0-4
shrinking, similar age brackets (20 to 54) start to
households. This belief usually translates into 10 8 6 4 2 0 2 4 6 8 10 10 8 6 4 2 0 2 4 6 8 10

misconceptions about women living in slums, as SLUM SLUM

grow, which probably reflects immigration from
it is assumed that low incomes and single moth- In Asia, an overview of selected countries reveals rural areas. This is particularly clear in the age
erhood go together. two findings: firstly, with the exception of Yemen, pyramid for South Africa. The age distribution of
women heading households tend to settle in urban Brazil and South Africa for the urban slum and
80+ 80+

However, UN-HABITAT data and analyses have areas as opposed to rural areas; and secondly, 75-79
70-74
65-69
75-79
70-74
65-69
non-slum population shows a significant demo-
60-64 60-64
shown that no clear pattern emerges on the mari- with the exception of Indonesia, the proportion of 55-59 Male Female 55-59 Male Female graphic shift: the age bracket from 0 to 14 years

Woman and child in Luanda, Angola EDUARDO LÓPEZ MORENO

50-54 50-54
45-49 45-49
tal status of women living in slums or about their women heading households in urban areas is larg- 40-44
35-39
40-44
35-39
for slum dwellers is as much as 20 per cent larg-
30-34 30-34
household responsibilities. The situation varies er in non-slum areas than in slum areas. Among 25-29
20-24
25-29
20-24
er than the same age distribution of the non-
15-19 15-19
from country to country. In some African coun- countries of the Commonwealth of Independent 10-14
5-9
10-14
5-9 slum population. Brazil exhibits a disproportion-
0-4 0-4
tries, such as Kenya, Malawi and Rwanda, States (CIS), such as Kazakhstan and Uzbekistan, 10 8 6 4 2 0 2 4 6 8 10 10 8 6 4 2 0 2 4 6 8 10 ately large prevalence of females under the age
woman-headed households are mostly found in NON-SLUM NON-SLUM of 19 years. This youth bulge is increasingly seen
the proportion of women heading households in
rural areas, whereas in Burkina Faso, Chad, as a major development challenge, since in the
cities is exceptionally high, 50 per cent.
Central African Republic, Egypt and Tanzania, the coming 5 years this group will enter young adult-
majority of woman-headed households are found hood and will be demanding housing and other
in slums, as opposed to non-slum and rural areas. What these statistics indicate is that the preva- 80+
75-79
80+
75-79

lence of woman-headed households, whether in

70-74
65-69
70-74
65-69 basic services as they marry, have children and
In a significant number of countries, including 60-64
Male Female
60-64
Male Female
55-59
50-54
55-59
50-54 form new households.
Cameroon, Côte d’Ivoire, Ethiopia, Ghana and rural, slum or non-slum areas is quite high. With 45-49
40-44
45-49
40-44
35-39 35-39
Morocco, the majority of woman-headed house- the exception of Africa, the share of such house- 30-34
25-29
30-34
25-29
20-24 20-24 When data is disaggregated between urban slum
holds are found within non-slum areas of cities. holds is also much greater in cities. 15-19
10-14
15-19
10-14
5-9
0-4
5-9
0-4 and urban non-slum, interpreting age pyramids
10 8 6 4 2 0 2 4 6 8 10 10 8 6 4 2 0 2 4 6 8 10 at simple glance becomes more complicated.
RURAL RURAL While comparing the above graphs of the urban
FIGURE 1.2.2 PROPORTION OF WOMAN-HEADED HOUSEHOLDS IN SELECTED COUNTRIES
slum and non-slum population, the most striking
BRAZIL SOUTH AFRICA finding is that the age pyramid of the slum popu-
60 lation has a clear expansive pattern, similar to a
conventional pyramid of a developing country. On
the other hand, the pyramid of the urban non-
50 The average slum dweller is younger and dies sooner than the slum population, while still maintaining a gener-
average non-slum dweller ally expansive pattern, tends to be relatively
40 constricted. This shows a trend toward reduction
Percentage (%)

in the younger age categories, which is typical of

Non-Slum UN-HABITAT urban indicators provide an 14), relative to the population as a whole is more developed countries.
30
All Slum overview of the distribution of the 0-4, 5-14 and higher. This is typical of an expansive pyramid
15-24 age groups in some of the countries that that has a wider base indicating that a large pro- Intra-city inequalities are clearly reflected in the
All Rural
20 are part of the organization’s global sample of portion of the population is young, which is also age-sex distribution of the slum and non-slum
cities. This distribution goes beyond the tradition- an indicator of high fertility rates. The rural pyra- populations in Brazil and South Africa. The two
al dichotomy of urban-rural, and presents a new mids of the two countries show that the propor- age pyramids serve to identify two clear trends:
10 breakdown into four categories: urban, rural, tion of people of the most productive age group an expansive broad base in the slum population
urban slum and urban non-slum. (20 to 54) shrinks. This may be attributed to structure, indicating a high proportion of chil-
0
rural-to-urban migration because of the greater dren, a rapid rate of population growth, and a
The above graphs of two middle-income coun- existence of employment opportunities in cities, low proportion of older people; and stable
co

so

da

ia

il

ti

ia

an

y
ga

az

ne

ke
an

bi

gu
an

Ha

d
Fa
oc

an

st
ne

In
Br

tries – Brazil and South Africa – show very con- which attract people in this age bracket. In coun-
r

growth in the non-slum population structure that

pi
Gh

ra
nz

kh

Tu
lo
or

Rw
a

ilip
Se

ca
in

Ta

Co

za
M

rk

Ni

Ph
Ka

ventional age pyramids for their respective urban tries with lower incomes, the proportion of rural
Bu

suggests a reduction of fertility rates and higher

and rural populations. The number of children (0- people in this age bracket shrinks further among life expectancy, respectively.
Source: UN-HABITAT 2005, Urban Indicators Programme, Phase III.
Note: Computed from Demographic and Health Surveys (DHS) data (1996-2004).
Source: UN-HABITAT 2005, Urban Indicators Programme, Phase III.

30 31
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Throughout Asia, slums are primarily the consequence of
only one shelter deprivation, reflecting fewer problems with
infrastructure and housing policy than in sub-Saharan Africa.
Therefore, governments or regional entities could tackle the
slum problem with simple sectoral interventions to improve
the living conditions of slum dwellers and their access to
secure tenure. For instance, in many sub-regions, one of the
main shelter deprivations identified is lack of access to ade-
quate sanitation; by addressing this shelter deprivation, coun-
tries can drastically reduce the number of slum dwellings.
Asia’s rapid economic growth over the last decade has not
made a significant impact on the elimination or reduction of
the urban inequalities on several fronts, and there are indica-
tions that disadvantaged groups remain disadvantaged, even
when economies improve.10 In fact, despite the region’s newly
industrializing countries, particularly China and India, which
rank among the world’s fastest-growing economies, gross
domestic product (GDP) rates have risen much more quickly
than national poverty rates have fallen, prompting the presi-
dent of the Asian Development Bank to refer to the scale of
deprivation in the region’s cities as “daunting”.11 It is interest-
ing to note that despite the sheer size of the problem, slum
dwellers in Asia have not received as much attention as they
warrant, although a recent Asia-Pacific Economic Cooperation
summit did emphasize the need to address urban poverty as a
priority issue.12
■ Slum trends in Latin America and the
Caribbean
Latin America and the Caribbean has almost completed its
urban transition; urbanization rates are stabilizing and slum
growth rates in the region are slowing down. The region’s
share of slum dwellers is 134 million people, less than the
total number of slum dwellers in just one country in Southern
Asia – India. In the 1960s and 1970s, Latin America’s expe-
rience with slum growth and slum prevalence was comparable
to that of the current situation in sub-Saharan Africa. This
trend was reversed in the late 1980s and early 1990s, when
the process of “re-democratization” resulted in the adoption
of progressive policies aimed at promoting more inclusive
governance and reducing inequalities. Notwithstanding these
positive developments, there is no room for complacency on
the part of policymakers. Extreme urban inequalities persist
throughout the region, and a considerably large share of slum
dwellers live on the edge of destitution. Although it is too
early to associate crime and violence levels with the number
of shelter deprivations in slum communities, many
researchers are beginning to make a link between inequality
and violence in the region’s larger cities.13
Argentina, Brazil and Mexico – the region’s three largest
countries – will be influential in reducing the proportion of
slum dwellers in the region by 2020. Brazil and Mexico have
already achieved a remarkable reduction in the share of slum
32
La Paz MARK EDWARDS/STILL PICTURES
Latin America and the Caribbean
has almost completed its urban
transition; urbanization rates are
stabilizing and slum growth rates
are slowing down.
dwellers within urban areas, while in absolute numbers, the
increase was minor in both countries. Slum growth rates are
very low in both Brazil and Mexico, at 0.34 per cent and 0.49
per cent per year, respectively. By 2020, the combined slum
population of Brazil and Mexico will have increased by only
MAP 3 URBAN POPULATION AND SLUM PROPORTION IN COUNTRIES OF LATIN AMERICA AND THE CARIBBEAN, 2001
4 million, totalling 71 million, if declining growth continues.
Argentina, however, is experiencing faster rates of slum
growth, at 2.21 per cent per year.
Low slum prevalence does not apply to all countries in the
region. In both Haiti and Nicaragua, more than 80 per cent
of the urban population lives in slums; in Bolivia, Guatemala
and Peru, slums host two-thirds of the urban population.
Bolivia, Guatemala, Nicaragua and Peru are all experiencing
high urban and slum growth rates, and slums with more than
one shelter deprivation are prevalent.
In the Caribbean country of Haiti, urban poverty levels and Urban Population
slum prevalence go hand-in-hand. The slum growth rate in <= 4 Mio
Haiti, 3.63 per cent per year, approximates slum growth pat- >4-10 Mio
terns in sub-Saharan Africa. Only one in four households has >10-30 Mio
a proper kitchen, and only 23 per cent of households have >30-75 Mio
access to improved water supply. Access to improved sanita- >75-150 Mio
tion is low, with only 16 per cent of all households, rich and >150 Mio
poor, having toilet facilities within the home. A substantial
proportion of households in the capital Port-au-Prince, 20 Percentage of
per cent, have only a pit latrine in the immediate vicinity. A Urban Slum Population
large proportion of the city’s garbage – 65 per cent – goes <=5%
uncollected. Many of the city’s families are at risk of being left
>5-10%
behind in an emergency, as 40 per cent of the dwellings do
not have a road leading to them and are not accessible by fire >10-40%
trucks or ambulances.14 However, the region as a whole >40%
stands out as one that has in recent years aggressively pursued
and implemented inclusive urban governance and slum
upgrading policies aimed at the most vulnerable populations. Source: UN-HABITAT, Global Urban Observatory 2005.
33
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

■ Degrees of shelter deprivation:
A regional analysis
Not all slums are homogeneous and not all slum dwellers suf-
fer from the same degree of deprivation. The degree of depriva-
tion depends on how many of the five conditions that define
slums (poor access to improved water, poor access to sanitation,
non-durable housing, insufficient living area and insecure
tenure) are prevalent within a slum household. UN-HABITAT
analyses show that sub-Saharan Africa’s slums are the most
deprived; over 80 per cent of the region’s slum households have
one or two shelter deprivations, but almost half suffer from at
least two shelter deprivations. Approximately one-fifth of slum
households live in extremely poor conditions, lacking more
than three basic shelter needs.
Generally, the lack of sanitation and water in the region’s
slums is compounded by insufficient living space for families
and inadequate, makeshift housing. One major reason families
have limited access to water and sanitation is that municipal
authorities often refuse to extend essential services to their
unplanned neighbourhoods. But the burden does not stop
there. When lack of sanitation is coupled with lack of water, or
temporary or overcrowded housing, the disease and labour bur-
den, especially on women and children, is even more intense.
This means that slum dwellers living under such hardship con-
ditions have to cope with survival issues for a considerable part
of each day. Many slum dwellers also lack secure tenure, which
makes their housing even more precarious. Slum formation and
growth is a complex problem to which African governments
must commit multiple sectoral interventions and investments
to lift their citizens out of poverty.
Slum households with the most shelter deprivations are high-
ly visible in most African cities, as many are clustered within
geographically contiguous high-density neighbourhoods, either
within or on the outskirts of cities. The concentration of slum
households is highest in Ethiopia, followed closely by Burkina
The most common deprivation experienced
by urban households in sub-Saharan Africa
is lack of access to improved sanitation;
45 per cent of the urban population suffers
from this deprivation, while 27 per cent
suffers from overcrowding.
Faso, Chad and Mali. In general, if a neighbourhood reflects
slum characteristics, so do most of the individual households
within that neighbourhood. This rule of thumb, however, does
not apply to all countries within the region. In Cameroon,
Ghana, Guinea, Namibia, and Zimbabwe, most slum house-
holds are individual dwellings in different neighbourhoods;
some also exist within serviced, middle- and high-income areas.
While sub-Saharan Africa’s cities suffer from the most
severe shelter deprivations, cities in Northern Africa have
managed to reduce the severity of slum conditions markedly;
a vast majority of slum households – 89 per cent – suffer from
only one shelter deprivation. Simple, low-cost interventions
in increasing access to improved sanitation for instance, are all
that are needed to help the countries in Northern Africa cre-
ate “cities without slums” – a goal they have been trying to
reach since long before the Millennium Declaration was
adopted in 2000.
Urban households in Southern Asia suffer from similar levels
of deprivation as those in sub-Saharan Africa. Nearly one in
five households lack two basic shelter needs. Among slum
households, one in three families has to cope with the lack of
two essential services. Nonetheless, unlike sub-Saharan Africa,
very few slum households in Southern Asia suffer from three or
more shelter deprivations, despite the huge magnitude of slums
within the sub-region, lack of sanitation and overcrowding are
the most common deprivations experienced by urban popula-
tions in the region. These figures may not reflect the reality on
the ground, however, as UN-HABITAT and official data
sources do not capture the shelter deprivations experienced by
pavement dwellers or street families, who are not normally cat-
egorized as “households” in censuses and surveys in countries
such as India.
Western Asian cities are similar to, albeit somewhat worse off
than, Northern African cities. Among slums, nearly one in four
lack more than two indicators of adequate shelter. The problem
in Western Asia is compounded by a volatile political situation,
which has contributed to an influx of refugees and internally
displaced persons (IDPs) to cities. In contrast, the majority of
slum households in South-Eastern Asia – 74 per cent – suffer
from only one shelter deprivation.
In Latin American cities, neither the magnitude of slums nor
the degree of severity is as daunting as in other regions.
However, the proportion of slum households that suffer from
at least one shelter deprivation is quite high: 66 per cent.

TABLE 1.2.3 & FIGURE 1.2.3 PROPORTION OF SLUM HOUSEHOLDS IN DEVELOPING REGIONS BY NUMBER OF SHELTER DEPRIVATIONS, 2001

Deprivations 100

Region One Two Three Four

80
Percentage

Percentage (%)
60
Northern Africa 89 11 0 0
Sub-Saharan Africa 49 33 15 3 40

Latin America and the Caribbean 66 25 8 1

Favelas in Rio De Janeiro JOHN MAIER

Eastern Asia - - - - 20

Southern Asia 66 29 5 0
0
South-Eastern Asia 74 20 5 1

Su rica n

& La frica n

be ica

So As rn

As rn

As rn
Af her

A ara

he

te
ut ia

ia

ia
rri er

st
an

es
rt

ut
Ca Am

Ea
Western Asia 77 16 6 1

Sa
No

W
So

h-
b-

th tin
e
four deprivations three deprivations

Source: UN-HABITAT 2005, Urban Indicators Programme, Phase III. two deprivations one deprivation

34 35
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6
M POPULATIO
D SLU N IN
WORL 202
3: 07
IO 00
AR m

EN

illi
SC

on
■ The prospect of reaching
the slum target
Forecasts reveal that the magnitude of slums will
continue to increase if the trends dominant
between 1990 and 2001 are projected into the
future. In light of recent evidence, even if govern-
ments collectively manage to improve the lives of
100 million slum dwellers by 2020 – as per the 1990 1995 2000 2005 2010 2015 2020
Millennium Development Goals and targets –
this achievement will be insignificant in relation
to creating “cities without slums”,15 a stated
objective of the Millennium Declaration.
Assuming that the leaders who developed the slum
target were aiming to address a major development
issue, policymakers should adjust the benchmark to
reflect the reality of slums of today and tomorrow.
In view of the existing slum situation around the
world, UN-HABITAT constructed three scenarios to 19
aid planners and policymakers who have a stake in 90 ON
improving the lives of slum dwellers. The worst-case sce-
:W
OR I LLI
nario (Scenario 1) assumes that the rate of slum growth
LD S
L U M P O P U L A TI O N 7 15 M
between 1990 and 2001 will remain the same in all five-year
periods between 2000 and 2020 – that is, slums will contin-
ue to grow. The second scenario (Scenario 2) assumes that FIGURE 1.2.4 MAGNITUDE OF SLUMS IN 2020 UNDER
there will be 100 million fewer slum dwellers in 2020 than THREE SCENARIOS
in 1990, which means the target will be met, but in an envi-
ronment in which the annual growth rate of slums exceeds Scenario 1
the rate at which they are being improved. The best-case sce- This assumes present trends of urban and slum growth continue
nario (Scenario 3) assumes that the proportion of slum unabated into the future.
SC
dwellers in 1990 will be reduced by half, in alignment with ENA ON
Scenario 2 RIO ILLI
most of the other Millennium Development Goals and targets.
2: W O .3 B
The result of the projection for Scenario 1 suggests that by Improve the lives of 100 million slum dwellers by 2020 i.e.
RLD SLU 2 020 1
2020, there will be nearly 1.4 billion slum dwellers in the
achieve MIllennium Development Goal 7, target 11. M P O P UL A TIO N IN
world, if present trends of urban and slum growth continue Scenario 3
unabated into the future, and if governments do not upgrade Reduce proportion of slum dwellers from 31 per cent of the global FIGURE 1.2.5
slums or provide positive alternatives to new slum formation. urban population in 1990 to 15 per cent in 2020. This would THE RISE AND RISE OF SLUMS, 1990-2020
Under Scenario 2, if the lives of 100 million slum dwellers are reduce by half the proportion of people living in slums.
improved, 1.3 billion people around the world will continue
400,000
to live in slum conditions. Scenario 2 reveals that slum
growth rates over the next 15 years will be highest in sub- In view of the general lack of international and national 350,000 Northern Africa
Saharan Africa. While the international community would commitment to achieving target 11 in most countries, the Sub-Saharan Africa
have achieved the very modest target set out in the best-case scenario seems quite elusive. If current trends con- 300,000 Eastern Asia
Millennium Development Goals, it would have made no sig- tinue, it is highly likely that in 2020 the slum population will 250,000
Eastern Asia excluding China
Population ('000)

nificant impact on reducing the proportion of people living be 1.4 billion. This has serious implications in terms of poli- Southern Asia
Latin America and the Caribbean
in slum conditions. Scenario 3 is based on the assumption cies for improving lives of slum dwellers, as well as the situ- 200,000
South-Eastern Asia
that in the next 15 years, international and national stake- ation of slum dwellers vis-à-vis the poverty, health, educa- Western Asia
150,000
holders will have adopted and implemented policies to pre- tion, and employment targets stated in the other Millennium Oceania
vent slum formation and reduce by half the number of slum Development Goals. The growth of slums in the world’s 100,000
dwellers. If these policies were effective, the share of slum cities, which will host the majority of the world’s population
50,000
dwellers would decline from 31 per cent of the urban popu- after 2007, should therefore be a cause for concern, as they
lation in 1990 to 15 per cent in 2020. This would amount to may eventually jeopardize the achievement of all the 0
improvement in the lives of at least 700 million people. Millennium Development Goals and targets. 1990 2001 2005 2010 2015 2020

Source: UN-HABITAT, Global Urban Observatory 2005.

36 37
 SCOTT LECKIE
THE SLUM TARGET IS NOT IN LINE WITH HOUSING RIGHTS
Children of pavement dwellers, Mumbai RASNA WARAH
Reviewing the Millennium Development Goals through a human rights
prism, one is forced to ponder whether these Goals were designed to
supplant, integrate or ignore human rights concerns of the world’s poor.
All fair-minded people, of course, would hope for the second of this tril-
ogy of options. Unfortunately, looking closely at the Goals, particularly
Goal 7, target 11 on improving the lives of slum dwellers, it appears the
Goals’ flaws, as far as housing rights are concerned, far out-number
their benefits.
Three shortcomings stand out as particularly worrying. First, the objec-
tive of improving the lives of at least 100 million slum dwellers by 2020
is almost obscene in its conscious exclusion of a huge majority of the
world’s urban and rural poor. In a world of almost one billion slum
dwellers, to speak of improving the lives of less than 10 per cent of the
world’s poorest citizens and rights-holders underscores just how far the
Goals stray from the language, sentiments and vision of human rights.
Which 100 million slum dwellers are we actually talking about? Who
will choose those whose lives will be improved? What say will they
have in the matter? Which 900 million or more slum dwellers and
homeless citizens will fall through the cracks? Who will inform them of
their plight of having been so deliberately barred from the Millennium
Development Goals? And what of the additional 400 million slum
dwellers that UN-HABITAT projects will be in need of improved hous-
ing by 2030? Does the world truly accept that perhaps some 1.3 billion
people will call slums home some fifteen years from now, and that this
is somehow an acceptable future for the planet’s urban poor?
Second, the Millennium Development Goals fail to address some of the
most pressing housing rights concerns affecting the world’s slum
dwellers, as if these, too, were somehow not part of the poverty trap
facing growing numbers of people. The global forced evictions epidem-
ic, decimated budgets and reductions in public expenditure on housing
for the poor, spiraling house, land and property prices reaching bubble-
like proportions in many countries, illegal land grabs, entrenched dis-
crimination against women, the disabled and the elderly, ethnic cleans-
ing, the demolition of homes during war and so many other core hous-
ing rights themes are all remarkably absent from the Goals. Not only
would addressing these poverty-expanding processes have been logi-
cal given their impact on hundreds of millions of dwellers throughout
the world, but by focusing on and forcefully discouraging these prac-
tices, the Goals would have improved the lives of many more millions
of slum dwellers than will possibly see improvements based on Goal 7,
target 11.
Third, the Goals – as with so many of the agreements emerging from
the various global summits during the past decade, whether by deci-
sion or by default – all too often end up taking the wind from the sails
of the human rights movement, slowing human rights progress and
shifting burdens of proof from national governments to the internation-
al community, which can never alone transform the human rights
dreams of the poor into reality.
The Millennium Development Goals are formulated as if to tease those
who – in precisely the same manner as international human rights law
– treat issues of poverty not only as development questions, but as
rights. By labeling what are, in fact, core human rights principles mere-
ly as “goals”, these internationally-agreed targets are effectively sup-
porting a creed which sees only half of the human rights equation as
actual human rights, somehow relegating the other half to “goals”,
“aspirations” or “needs” – not enforceable rights held and rightfully
expected by all, especially those currently without the protection these
rights are meant to provide.
As far as housing rights are concerned, it would be difficult not to con-
clude that the Millennium Development Goals let governments off the
hook; they almost insinuate that a staggeringly large portion of human-
ity is condemned by circumstance to live in life- and health-threatening
conditions, without security, as if to say, “Yes, we care, but as far as
slums dwellers go, we know and accept that our care will only reach a
few of you.” Good luck to the 900 or so million slum dwellers that the
Millennium Development Goals forgot about. They will need it. In the
coming years, these neglected millions will, as always, gain ground,
organize, and support themselves. Whatever rights slum dwellers
accrue, or security they can claim, will come from their own energies
and sadly, not from a global accord agreed to by governments.
Scott Leckie is the Executive Director of the Centre on Housing Rights
and Evictions (COHRE).
Endnotes
1 Milanovik 2005.
2 Definition used in Concise Oxford English Dictionary. Tenth Edition, 2002.
3 UN-HABITAT slum estimations revised in 2005, based on coordination
of definitions with the WHO/UNICEF Joint Monitoring Programme for
Water Supply and Sanitation.
4 Only data for the developing world, collected with household surveys
and censuses, are reported here, as UN-HABITAT’s estimates for the
developed world are based on secondary sources, including reports,
printed censuses and modeling.
5 UN-HABITAT 2004b.
6 Ibid.
7 The Orangi Project, which started as a pilot project within a township
in Karachi in 1965, has been touted as one of the most successful
demonstration projects in the world and has since been replicated in
seven cities in Pakistan.
8 Dialogue at Second World Urban Forum, Barcelona, September 2004.
9 In China, people’s status as residents is determined by their ability to
acquire residency permits for a particular city, town or village. An inde-
pendent study conducted by the Fafo Institute for Applied International
Studies in Beijing in 2000 attests to the fact that the population that
comes either from other cities or villages or from the rural areas of
Beijing is not considered part of the total city population. Since they
are not regarded as dwellers of the city, they are not considered slum
dwellers of the city either. However, the study indicates that those
who belong to settlements outside Beijing’s urban areas constituted
16 per cent of the de facto population of Beijing at the time. Most of
the slum characteristics that UN-HABITAT uses as criteria are preva-
lent among this group of people. The share of people originating from
outside villages could be even higher, as many migrant workers do not
live within established housing, as they do not have permits to live in
cities - many live on construction sites. Quite a substantial proportion
of these inhabitants suffer from overcrowding; they often live in dor-
mitories, jerry-built houses or workers’ huts, shared apartment units or
simple apartment units with shared facilities.
10 Forum on the Global South 2003.
11 Asian Development Bank President Haruhiko Kuroda quoted in
Newsweek, 21 November 2005.
12 The Asia-Pacific Economic Cooperation (APEC) summit held in Busan in
November 2005.
13 For instance, the 2005 United Nations report entitled The Inequality
Predicament shows a strong correlation between inequality in Latin
American cities and levels of homicide.
14 Fafo Institute for Applied International Studies 2003.
15 UN Millennium Declaration 2000, paragraph 19.
39
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

1.3 How Well is Your Country Performing on the Slum Target?

rates over the last 14 years. Yet, these countries still have large
numbers of people lacking adequate housing and basic services –
Millennium Development 52 million in Brazil, 22 million in Indonesia, 15 million in
Goal 7, target 11: Mexico and 8 million in South Africa. These countries are on the
A Global Scorecard By 2020, to have achieved a
significant improvement in the lives
right path but clearly have some way to go in order to make the
kind of reductions registered by the “on track” countries.
of at least 100 million slum dwellers
Warning signals light up in “at risk” and “off track” countries
■ A snapshot of country performance The worrying trend is that most developing countries have
failed to make much headway in reducing slum growth rates. Of

D
eveloping countries have pursued a range of particular concern are those countries that experienced substan-
countries using statistical data and internationally-agreed defini-

Laying new water pipes in Mathare Valley, Nairobi FRIEDRICH STARK/STILL PICTURES
policies and practices to deal with the deficien- tial slum growth rates (ranging between 4 per cent and 6 per cent
tions and methods.
cies in the provision of basic services, housing, annually) and high incidence of slums in the last 15 years. This
But things are changing. The movement that is now challeng-
health, and education for the urban poor and combination has had lethal effects. For instance, the slum popu-
ing countries and the international community to live up to the
slum dwellers over the course of the last thirty lation in Tanzania has more than doubled in the last 15 years,
promise of the Millennium Development Goals is helping to trig-
years. The international community, including United Nations from 5.6 million in 1990 to 14 million in 2005. Likewise in
ger a new way of measuring the performance of countries in
agencies, multilateral funding agencies and bilateral donors, Bangladesh, Nigeria and Sudan, the numbers of slum dwellers
improving slums and meeting the other Goals and targets in
have, through successive strategies, arguably played a pivotal grew from 19 million, 24 million and 5.7 million in 1990 to 36
urban areas. International agencies are indeed seizing the oppor-
role in transforming government attitudes and policy respons- million, 46 million and 12 million in 2005, respectively. This “off
tunity to produce more and better statistics with which to mon-
es to slums. The policy options and interventions by both the track” group needs to take drastic action now to improve existing
itor the Goals.2 UN-HABITAT has devised, through extensive
international community and governments have changed over slum conditions and prevent future slum formation; otherwise
consultation with its partners, a new methodology for measuring
time: neighbourhood-level self-help solutions to housing and the numbers of slum residents will continue to rocket upwards.
slums and has subsequently produced estimates for the numbers
in situ upgrading during the late 1970s and 1980s; getting the Other countries, including Argentina, China, India and Morocco
of slum dwellers at global, regional, national and city levels.3 The
“enabling environment” right and improving urban manage- are doing slightly better than this group in terms of managing
results have been used in this part of the Report to construct a
ment in the 1990s; and scaling-up of slum upgrading through slum growth rates at roughly 2 per cent a year, but they are still
global scorecard showing the varying performance of over 100
national and citywide programmes since the end of 1990s. in the “at risk” category of countries, as the proportion of people
countries in improving the lives of slum dwellers and reducing
During this period, such remedies have brought hope to living in slum conditions is relatively high (over 30 per cent) and
slum growth rates.
many of the world’s urban poor. Initiatives such as the they still need to revisit existing policies and improve perform-
Kampong Improvement Programme in Indonesia, the Favela ance. Over 70 per cent of the countries analysed fell under these
Bairro programme in Rio de Janeiro, Brazil, and the Million ■ Results at a glance two bottom-end performance categories.
Houses Programme in Sri Lanka have significantly improved
the lives of the urban poor. For example, the programme in Countries “on track” – those starting to make urban
Indonesia managed to reach 15 million people over its 30- poverty history ■ Track record of regions
year history working with some 300 local governments in the Among all developing countries, Thailand has seen the sharpest
provision of water, sanitation, shelter and roads.1 decline in slum growth rates. In 1990, there were almost 2 mil- Moving from a country to a regional portrait, the scorecard
And yet for much of the rest of the developing world, pol- lion people living in slum conditions; by 2005, this figure had shows countries in sub-Saharan Africa struggled above all to cope
icy reforms or interventions have not been enough or simply been slashed to just 119,000. This dramatic decrease is attributed with the rising numbers of slum dwellers – 34 out of the 50
failed to materialize. Despite their good intentions, some gov- largely to the government’s long-standing commitment to imple- countries in the “off track” group are in this part of the world,
ernments and donors have struggled to cope with overwhelm- menting programmes to improve the housing conditions of the including, Kenya, Lesotho and Mali. These countries not only
ing demographic pressures, massive backlogs in basic services urban poor. Egypt, Georgia, Sri Lanka and Tunisia are also experienced some of the highest slum growth rates, but also tend-
and housing provision, and growing environmental degrada- among the “on track” countries registering falling slum growth ed to have a large proportion of their total urban population liv-
tion and unemployment, while other governments continue rates since 1990, and are seeing significant reductions in the ing in slums. South Africa is the only country in the region that
to ignore the issue of slums in official policy circles. When number of people living in slums. Egypt succeeded in reducing made it into the higher “stabilizing” category, recording an
remedies fail to reach people they are meant to serve, it is the the number of slum dwellers by 3 million from 1990 to 2005. almost negligible annual slum growth rate.
poor who fall back on their own capacity and resilience to Tunisia has succeeded in more than halving the number of slum Southern Asia, as a whole, also demonstrated a poor track
make a home and living for themselves in the city. dwellers in the same period to approximately 190,000. However, record. In particular, Bangladesh, Nepal and Pakistan continued
However, it is still not very clear where actions to upgrade these countries are among a meagre 14 countries out of the over to lag behind, while India performed somewhat better in manag-
slums or prevent their formation have collectively made the 100 analysed that made it to the “on track” category. ing overall slum prevalence, with a slum growth rate of 1.7 per
biggest difference in improving lives of the urban poor or cent. As a notable exception, Sri Lanka, following decades of
where they have failed to address the problem of slums. While Countries in the “stabilizing” category – those starting to put social investment, ranks among the top performers not only in
for the past three decades, researchers and practitioners have the brake on slum growth rates, but need to monitor and the region but also worldwide; with an annual decline of 3.7 per
produced mostly anecdotal or qualitative evidence to evaluate make sure they don’t slip back cent, the total number of slum dwellers stands at half a million,
progress and failure of countries and cities in slum improve- Brazil, Indonesia, Iran, Mexico, South Africa and Turkey appear down from nearly 900,000 in 1990.
ment, what has been missing is a much more systematic, rig- to have made fairly good progress in basic service provision to the Eastern Asia demonstrates even more mixed results, but averages
orous attempt to compare performance among and within urban poor, which is reflected in low to almost stable slum growth a better overall performance than Southern Asia. While Thailand

40 41
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

■ Scorecard Methodology ON TRACK STABILIZING AT RISK OFF TRACK

Slum Slum Slum Slum Slum Slum Slum Slum Slum Slum Slum Slum
The chart presents the first findings of the global scorecard on slums. Annual % % Pop. Pop. Annual % % Pop. Pop. Annual % % Pop. Pop. Annual % % Pop. Pop.
growth slum slum 1990 2005 growth slum slum 1990 2005 growth slum slum 1990 2005 growth slum slum 1990 2005
Countries are judged to be doing well if they have managed to reduce or reverse rate 1990 2005 (000s) (000s) rate 1990 2005 (000s) (000s) rate 1990 2005 (000s) (000s) rate 1990 2005 (000s) (000s)
slum growth rates and if, at the same time, they have succeeded in keeping the
proportion of slum dwellers at relatively low levels.
The cut off points for the annual slum growth rates were: 0%, > 0-2%, > 2-

UPPER MIDDLE INCOME*

4%, > 4%. These growth rate categories were combined with the proportion of Puerto Rico 1.6 2.0 2.0 50 63 Brazil 0.3 45.0 34.0 49,806 52,374 Argentina 2.2 30.5 34.1 8,597 11,978 Gabon 5.9 56.1 70.4 357 872
slums in the discussed regions. As the incidence of slums is very different in Uruguay -10.3 6.9 1.3 191 41 Mexico 0.5 23.1 18.5 13,923 14,983 Botswana 3.7 59.2 61.3 311 540 Oman 5.4 60.5 60.5 671 1,506
Trinidad and Tobago 0.5 34.7 31.0 292 317 Costa Rica 4.3 11.9 13.1 195 372 Venezuela 2.5 40.7 40.7 6,664 9,642
these regions, different categories for the slum proportions were chosen to char- Lebanon 3.1 50.0 50.0 1,142 1,811
acterise these different conditions. For Africa and Asia, where the incidence of Panama 2.2 30.8 30.8 397 552
slums is generally higher than in the other regions, the cut-off points were the Saudi Arabia 3.8 19.8 19.8 2,385 4,196
same (30%, >30-60%, >60%). For Latin America and the Caribbean, the cut
off points were 10, >10-40, 40% and for the Commonwealth of Independent
States the categories are 20, >20-50, >50.
As a result of this procedure the countries could be grouped within the four Cuba 0.7 2.0 2.0 156 174 Colombia 1.1 26.0 20.5 6,239 7,381 Algeria 3.0 11.8 11.8 1,508 2,370 Bolivia 2.3 70.0 58.4 2,555 3,597

LOWER MIDDLE INCOME*

Egypt -1.6 57.5 34.9 14,087 11,015 Dominican Republic -0.9 56.4 32.4 2,327 2,038 China 2.3 43.6 35.9 137,929 195,682 Guatemala 2.5 65.8 60.4 2,192 3,186
performance categories. Covering over 100 countries in the developing world, Sri Lanka -3.7 24.8 10.9 899 515 El Salvador 1.9 44.7 32.3 1,126 1,495 Ecuador 2.5 28.1 24.8 1,588 2,317 Peru 3.4 60.4 71.1 8,979 14,862
the scorecard measures trends in country performance between 1990 and 2005. Thailand -18.8 19.5 0.9 1,998 119 Iran, Islamic Rep. of 1.6 51.9 41.6 17,094 21,763 Honduras 2.4 24.0 16.3 488 703
Countries are grouped into the following four performance categories: Tunisia -5.4 9.0 2.7 425 188 Paraguay 0.5 36.8 21.7 756 812 Iraq 2.5 56.7 56.7 6,825 9,992
Philippines 1.9 54.9 40.7 16,346 21,792 Jamaica 3.5 29.2 38.4 356 604
ON TRACK South Africa 0.2 46.2 29.4 8,207 8,439 Jordan 4.3 16.5 15.4 388 741
Turkey 0.0 23.3 16.3 7,997 8,016 Morocco 2.0 37.4 31.1 4,457 6,054
Countries experiencing rapid, sustained decline in slum growth rates in Kazakhstan -0.6 29.7 29.7 2,835 2,605 Namibia 2.9 42.3 36.4 155 239
urban areas and/or those with low slum prevalence. Syrian Arab Republic 3.2 10.4 10.4 629 1,012

STABILIZING Georgia -7.2 18.4 6.4 558 189 Indonesia 1.4 32.2 20.5 17,964 22,049 Bhutan 1.2 70.0 37.3 61 73 Afghanistan 6.4 98.5 98.5 2,458 6,375
Countries starting to stabilize or reverse slum growth rates but which need Myanmar 1.3 31.1 24.9 3,105 3,794 Congo 3.6 51.9 48.7 5,366 9,227 Angola 5.3 83.1 83.1 2,193 4,839
Republic of Moldova -1.3 31.0 31.0 634 522 India 1.7 60.8 53.6 131,174 169,671 Bangladesh 4.3 87.3 83.8 18,988 36,079
LOW INCOME*

to monitor progress to ensure sustained reductions. Mongolia 0.7 68.5 63.6 866 969 Benin 5.3 80.3 84.8 1,288 2,870
Viet Nam 1.2 60.5 43.4 8,100 9,632 Burkina Faso 4.0 80.9 74.9 987 1,791
AT RISK
Burundi 2.7 83.3 59.8 294 438
Countries experiencing moderate to high slum growth rates but also having Cambodia 6.1 71.7 72.4 870 2,162
moderate incidence of slums that require remedial policies to reverse growth Cameroon 5.0 62.1 68.9 2,906 6,197
Central African Rep. 3.1 94.0 91.8 1,038 1,646
in numbers of slum dwellers. Chad 4.3 99.3 99.1 1,218 2,308
Congo, Dem Rep. of the 5.2 84.5 92.2 1,050 2,276
OFF TRACK
*WORLD BANK INCOME GROUP Côte d’Ivoire 6.0 50.5 75.6 2,532 6,203
Countries with already high slum proportions, facing rapid, sustained slum Eritrea 3.6 69.9 69.9 342 590
growth rates and which require immediate, urgent action to slow down or FIGURE 1.3.1 A GLOBAL SCORECARD ON SLUMS 2006 Ethiopia 4.8 99.0 99.5 5,984 12,315
Gambia 5.4 67.0 67.0 155 348
reverse slum trends. Source: UN-HABITAT 2005, Urban Indicators Programme, Phase III. Ghana 1.8 80.4 66.0 4,083 5,372
Guinea 3.4 79.6 69.8 1,145 1,918
Guinea-Bissau 5.2 93.4 93.5 210 456
Haiti 3.6 84.9 86.0 1,728 2,976
MAPS 4, 5, 6 A LOOK AT COUNTRY PERFORMANCE IN DEVELOPING REGIONS, 2006 Kenya 5.9 70.4 70.9 3,985 9,620
Lao People's
Dem Republic 4.7 66.1 66.1 422 850
Lesotho 6.3 49.8 59.8 168 434
Liberia 2.0 70.2 51.2 632 853
Madagascar 5.3 90.9 93.7 2,562 5,696
Malawi 3.9 94.6 89.9 1,033 1,860
Mali 4.9 94.1 92.9 1,968 4,083
Mauritania 5.6 94.3 94.3 827 1,915
Mozambique 6.9 94.5 93.9 2,722 7,710
Nepal 4.8 96.9 90.9 1,574 3,213
Nicaragua 3.4 80.7 81.0 1,638 2,730
ON TRACK Niger 5.9 96.0 96.3 1,191 2,882
Nigeria 5.0 80.0 71.9 24,096 46,272
STABILIZING Pakistan 2.7 78.7 71.8 26,416 39,722
AT RISK Rwanda 3.5 82.2 90.1 296 504
Senegal 4.1 77.6 76.0 2,276 4,181
OFF TRACK Sierra Leone 3.6 90.9 97.6 1,107 1,895
Somalia 3.6 96.3 97.4 1,670 2,867
NO OR NOT RELIABLE DATA Sudan 5.2 86.4 85.5 5,708 12,441
Togo 4.3 80.9 80.5 796 1,510
Tanzania, U. Rep. of 6.2 99.1 89.6 5,601 14,113
Uganda 5.3 93.8 92.7 1,806 4,010
Yemen 5.0 67.5 64.2 1,787 3,803
Zambia 2.9 72.0 74.8 2,284 3,519
Source: UN-HABITAT, Global Urban Observatory 2006.

42 43
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

growth rates. A detailed explanation as to why some countries – that of scaling up improvements that will reach the large

Businesswoman in Tunisia VINCENT KITIO

appears to be very much in a league of its own, Indonesia and
the Philippines have succeeded in keeping slum growth rates
low. China, despite impressive policy interventions, recorded a
relatively large slum population. Preliminary UN-HABITAT
analysis attributes this to high prevalence of poor sanitation in
cities; a WHO/UNICEF assessment in 2000 showed that 33.8
per cent of the urban population lacked access to improved san-
itation. Cambodia struggled to stay ahead of demographic and
poverty curves during the 1990s and experienced a high annu-
al slum growth rate of 6.1 per cent, resulting in a more than
doubling of the slum population from 870,000 in 1990 to over
Egypt and Tunisia are the “high-flyers” in
Northern Africa with a good track record of
investments in water and sanitation, as well
as informal settlements upgrading.
The best-performing developing region appears to be
Latin America and the Caribbean. The majority of countries
mainly fall under the “at risk” and “stabilizing” groups. The
top performers, according to the scorecard, are Cuba and
Uruguay. On the other hand, urgent attention is needed to
deal with overwhelming demand for services in countries
such as Bolivia, Guatemala, Haiti, Nicaragua, Peru and
Venezuela. However, the majority of countries in the region,
including Brazil, Colombia, Costa Rica, Honduras and
Mexico, appear to be coping more effectively and have seen
relatively low to moderate increases in the numbers and pro-
succeed in reducing slum growth and preventing slum forma-
tion, while others struggle to deal with growing poverty and
inequality in their cities is provided in Part Four of this Report.
The analysis draws on recent data compiled and generated by
UN-HABITAT to compare country performance and national
policy reforms in slum upgrading and prevention.4
Based on the scorecard and policy analyses carried out by
UN-HABITAT, we now have a better understanding of what
drives a country’s performance in improving the living con-
ditions of slum dwellers. We know that political commit-
ment and long-term government policies are essential ingre-
dients. The experience of some of the best-performing coun-
tries shows that there is nothing like the commitment of top
political leadership to give clarity of purpose, direction and
a sense of urgency in tackling head-on the growth of slums –
it has often proven to be the surest way of committing
actions and resources to the problem. In Morocco, when the
King declared slum upgrading as one of his top four priori-
ties, more than 40 per cent of the budget of the national
development plan went to upgrading. Brazil’s urban pro-
poor policies were given a boost when the President set up a
housing fund of $1.6 billion for house building and favela
upgrading. And in Tunisia, scaled-up upgrading has been a
core business of the government for three decades and is
consistently included in its successive five-year national
development plans.
“On-track” countries have also effectively hit the most crit-
ical policy levers, carrying out reforms and scaling-up of slum
upgrading programmes that have led directly to improvement
in the lives of slum dwellers. We know that some of the most
progressive sectoral reforms that drive better performance
focus on: improving secure land tenure; setting up a proper
land regulation system; providing affordable and accessible
housing; and improving the efficiency and coverage of water
numbers of people who are living in slums today, as well as pre-
venting future slums. This experience is something that other
struggling nations and cities could adapt, but they may not find
it easy in conditions where economic development, good local
governance and political leadership are weak.
■ Using the scorecard:
Measuring, managing and motivating change
The scorecard presented in this Report is only the first step
towards a better understanding of which countries are on track
to meet target 11 and why they appear to be more successful
than others. It does not pretend to provide definitive answers to
many of the questions indicated above. There is still a long way
to go to further develop the scorecard approach, especially in
terms of refining the outcome performance measures for slum
growth, producing more complete information about policies
for countries, and setting up a more rigorous way of rating and
comparing policy performance.
The scorecard should be seen as more than just a measuring
yardstick for measuring progress; rather, governments, civil
Political commitment to slum
prevention and upgrading is essential
to reducing slum growth.

Cairo VINCENT KITIO

2 million in 2005.
In largely Arabic-speaking Northern Africa and Western Asia,
the results have been mixed: while the former performed well on
reducing slum growth rates, the latter had higher slum growth
rates than those of the rest of Asia. Slum growth in many
Western Asian countries is not necessarily confined to lower-
income countries, but also extends to middle-income and high-
income countries in the region, such as Iraq, Jordan and Saudi
Arabia, which experienced a drop in economic performance in
the 1990s and witnessed sustained growth in the number of slum
dwellers (3 per cent to 4 per cent a year), from 6.8 million,
390,000 and 2.4 million in 1990 to 10 million, 740,000 and 4.2
million in 2005, respectively. Mauritania and Yemen also per-
formed poorly, recording annual slum growth rates of 5 per cent
or more. In contrast, Egypt and Tunisia are the “high-flyers” in
Northern Africa with a good track record of investments in water
and sanitation, as well as informal settlements upgrading, during
the course of the 1990s that resulted in a significant decline in
the number of slum dwellers, from 14 million and 425,000 in
1990 to 11 million and 188,000 in 2005, respectively.
portions of slum populations.
■ Understanding the dynamics of improved
performance
Six years after the Millennium Declaration committed gov-
ernments to strive for “cities without slums”, the results of the
global scorecard suggest that only a handful of countries in the
developing world have made real progress in providing better
public services and housing to slum dwellers. “On track” coun-
tries, where slum growth rates have actually fallen, include
Egypt, Sri Lanka, Thailand and Tunisia. In these countries,
governments and other stakeholders have taken on the respon-
sibility of providing decent housing, water and sanitation to
slum dwellers – and they have shown how to get the job done.
These countries also provide a beacon of hope and inspiration
to their neighbours. However, the worrying trend is that most
developing countries, particularly in sub-Saharan Africa, failed
to make much headway in reducing or reversing the slum
supply and sanitation services.
Some of the better performing countries also score highly
on local governance, accelerating effective decentralization
policies, municipal reforms and broad-based participation in
planning. With new-found mandates and powers and increas-
ing pressure to respond to voters’ demands, municipalities are
paying more attention to improving slums and are getting
more involved in slum upgrading. The analyses also show
that, in many cases, there are no hard-and-fast rules regarding
which policy interventions or governance structures work
best. For instance, some of the best performing countries,
including Egypt, Russia and Tunisia, have succeeded in
reversing slum growth rates under very centralized systems of
governance, while others, such as Colombia and India, have
been unable to reverse slum growth despite highly decentral-
ized governance structures.
The evidence also points to the high performers improving
the financial sustainability of local governments, and finding
innovative ways of raising domestic investment. We also know
that the most successful countries face up to a double challenge

44 45
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
society and the international community have an opportunity
Slum Housing in Manila, Phillipines A. APPELBE/UNEP/STILL PICTURES
to use the results to better manage their performance. By com-
paring the performance outcomes of different slum policies
among countries, governments and international development
agencies, stakeholders can use the results to improve policy
and resource allocation decisions, drive performance improve-
ments in service delivery for the urban poor, and communicate
to the public how well or otherwise progress is being made
towards target 11 and the other Millennium Development
Goals and targets in cities. More specifically, the scorecard
could be used to:
• Identify best practices in slum upgrading and prevention
using reliable statistical information. The scorecard may
help point toward success stories in improving the lives of
slum dwellers, and provide a basis for analysing what factors
led to better performance. Other countries could adapt and
implement the most effective practices in order to achieve
similar results.
• Motivate countries to improve their performance. One of
the most critical barriers to effective service delivery in the
public sector is the lack of competition – comparing and
reporting on performance can help demonstrate how well
agencies are doing compared with others.5 The scorecard
could foster a competitive spirit among countries, particu-
larly those at a similar level of development, motivating
progressive countries to make further, continuous
improvements and to alert lagging countries to take
urgent, remedial action.
• Strengthen the accountability of governments.
Ultimately, the performance information contained in
the scorecard, when made publicly available in countries,
could help to make governments more accountable to the
urban poor. While governments do not have the primary
responsibility of providing basic services, a responsibility
that is often shared with non-governmental organiza-
tions, the private sector and the urban poor themselves,
such a scorecard can lead to wider public debate and cre-
ate pressure for change.
• Help the international community to target aid more
effectively. The scorecard could provide donors, multilater-
al funding agencies and the United Nations system with a
tool for comparing how well countries are performing on
target 11. Results could be used to identify where problems
exist and to set priorities for resource allocation according
to level of need. The scorecard could also provide a baseline
against which future changes can be tracked. If there is evi-
dence of sustained lack of effective policies to bring down
slum growth, the international community, including
donors, could use the scorecard to decide if making any fur-
ther investments in slum upgrading or prevention to non-
performing countries is worthwhile.
46
The real energy behind this tracking system should come, not
from the donors, but from the countries that fall under the
spotlight of the scorecard. It is their policies, their political
leaders and their future prospects that are under scrutiny. The
issue is whether the results will grab the attention of politicians,
policymakers and the general public to produce sufficient pres-
sure on governments for change. Arguably, indicators and sta-
tistics make headlines if they are about issues that people really
care about and feel that they, or their leaders, have the power to
change for the better. It is too early to say if target 11 can trig-
ger such fundamental changes in attitude – but it is clear that
the momentum behind this movement is growing.
Progressive sectoral reforms, such as
improving land tenure and regularization,
providing affordable housing and
improving coverage of water and
sanitation, are key to slum prevention.
Endnotes
1 MIT website, “Upgrading urban communities”:
http://web.mit.edu/urbanupgrading.
2 Moreno 2005.
3 Refer to UN-HABITAT 2003c.
4 The scorecard simply shows where and by how much there has been
progress or otherwise in reducing the numbers of people living in slums.
In order to investigate the reasons behind good and bad performance,
the results were used to trigger an in-depth analysis of the policy envi-
ronment in 23 of the over 100 countries broadly representative of the
four performance categories. The 23 countries analysed were: Cuba,
Egypt, Sri Lanka, Thailand, and Tunisia in the “on-track” category; Brazil,
Colombia, Indonesia, Mexico, Philippines, and South Africa in the “sta-
bilizing” category; Chile, Haiti, India, and Morocco in the “at-risk” cate-
gory; and Afghanistan, Bangladesh, Burkina Faso, Ethiopia, Ghana,
Liberia, Senegal, and Tanzania in the “off track” category.
5 Morley, et al. 2001.
47
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
1.4 The Struggle to Achieve the Millennium Development Goals
will be Won or Lost in Cities
Singapore ©PHIL DATE. IMAGE FROM BIGSTOCKPHOTO.COM
THE GOOD NEWS
Cities drive national economies
T
he global fight against poverty1 – encapsulated in
the Millennium Development Goals2 – is heavi-
ly dependent on how well cities perform. The
link between urbanization and socio-economic
development cannot be disputed. Cities make
countries rich. Countries that are highly urbanized have higher
incomes, more stable economies, stronger institutions and are
better able to withstand the volatility of the global economy
than those with less urbanized populations.3 The experiences of
developed and developing countries also indicate that urbaniza-
tion levels are closely related to levels of income and perform-
ance on human development indicators.
Cities around the world are playing an ever-increasing role in
creating wealth, enhancing social development, attracting
investment and harnessing both human and technical resources
for achieving unprecedented gains in productivity and compet-
itiveness. As countries develop, urban settlements account for a
larger share of national income. In both developed and develop-
ing countries, cities generate a disproportionate share of gross
domestic product (GDP)4 and provide huge opportunities for
investment and employment.
Urban-based economic activities account for up to 55 per
cent of gross national product (GNP)5 in low-income countries,
73 per cent in middle-income countries and 85 per cent in high-
income countries.6 In the United States, for example, some
48
cities outpace even some countries in economic output. If the
five largest cities in the United States – New York, Los Angeles,
Chicago, Boston and Philadelphia – were treated as a single
country, it would rank as the fourth largest economy in the
world.7 This trend is also evident in the developing world: São
Paulo, Brazil’s largest city, and Bangkok, the capital of Thailand,
both host just over 10 per cent of the total population of their
respective countries, but both account for more than 40 per cent
of their countries’ GDP. Cities also generate a disproportionate
amount of revenue for governments; the residents of India’s
commercial capital Mumbai, for instance, pay almost 40 per
cent of the nation’s taxes.8
Goods and services are generally produced more efficiently in
densely populated areas that provide access to supportive servic-
es, transport and communication links, a pool of labour with
appropriate skills, and a critical mass of consumers – all attractive
qualities associated with cities. In the new, increasingly knowl-
edge-based global economy, cities are particularly efficient pro-
ducers. Improved economic and social infrastructure, together
with economies of scale and agglomeration benefits associated
with large urban centres, allow businesses and enterprises in cities
to flourish. The concentration of economic activity in cities
makes them prime generators of non-agricultural employment in
both the formal and informal sectors. While the formal sector
accounts for a much larger share of urban employment in indus-
trialized countries, the informal sector employs a significant pro-
portion of the non-agricultural labour force in developing
regions (up to 80 per cent in sub-Saharan Africa and more than
60 per cent in Asia and Latin America).
Cities are also engines of rural development. They provide
many opportunities for investment, which not only support
FIGURE 1.4.1 HUMAN DEVELOPMENT IS CLOSELY RELATED TO urban development but also contribute to rural development in
LEVELS OF URBANIZATION an environment of strong urban-rural linkages. Improved infra-
1.000
structure between rural areas and cities increases rural produc-
tivity and enhances rural residents’ access to education, health-
0.900
care, markets, credit, information and other services. On the
0.800
other hand, enhanced urban-rural linkages benefit cities
0.700
through increased rural demand for urban goods and services
Score of HDI 2003
0.600 and added value derived from agricultural produce. Increased
0.500 productivity and competitiveness also fuels the urbanization
0.400 process: all over the world there are examples of sleepy fishing
0.300
villages becoming thriving ports, barren outposts becoming
major trading centres and railway depots or harbours becoming
0.200
capital cities. Urban transformations often translate into positive
0.100
performance on human development indicators and reduced
0.000 poverty in both rural and urban areas. Put together, all of these
0 20 40 60 80 100
factors provide an apt environment for the attainment of the
Percentage urban by country 2003
Millennium Development Goals and targets.
Sources: United Nations, World Urbanization Prospects: The 2003 Revision, UNDP,
However, the relative absence of infrastructure, such as roads,
Human Development Report 2005.
Note: Developing countries with populations of more than one million are plotted. water supply, communication facilities, and adequate housing in
The human development index (HDI) is a summary measure of human development small- and medium-sized cities – which are currently absorbing
that measures the average achievement in a country in three basic dimensions: a most of the world’s urban population growth – makes these
long and healthy life, as measured by life expectancy at birth; knowledge, as
cities less competitive at the national, regional and global levels.
measured by the adult literacy rate and the combined primary, secondary and terti-
ary gross enrolment ratio; and a decent standard of living, as measured by GDP In many countries, a disproportionate amount of public invest-
per capita (PPP US$). ment, especially investment in infrastructure, goes to the larger
Countries that are highly urbanized have
higher incomes and more stable
economies than those with less urbanized
populations.
cities, particularly national capitals. Attempts to “decentralize”
economic activities to secondary cities are unlikely to be success-
ful unless the decentralization is supported by pro-poor invest-
ment in infrastructure and public services, and by the financial
and institutional strengthening of local authorities.9
Contrary to popular perception, infrastructure investments
in urban areas are not only cost-effective but also environmen-
tally sound. The concentration of population and enterprises in
urban areas greatly reduces the unit cost of piped water, sewers,
drains, roads, electricity, garbage collection, transport, health
care, and schools. However, the cost-effectiveness of infrastruc-
ture investment is greatly reduced when these investments are
made too late. For instance, when informal settlements or
slums are allowed to proliferate, it becomes more difficult and
more expensive to install infrastructure and services because no
prior provision was made for the settlement’s development.10
Moreover, population densities and the spatial configuration of
slums often do not allow for the development of roads, sewer-
age systems and other facilities that may be easier to install in
less dense and better-planned areas.
The Millennium Development Goals provide an apt frame-
work for linking the wealth of cities with increased opportuni-
ty and improved quality of life for their poorest residents. In
many countries, however, prosperity has not benefited urban
residents equally. Mounting evidence suggests that economic
growth in itself cannot reduce poverty or increase opportuni-
ties if it is not accompanied by equitable polices that allow
low-income or disadvantaged groups to benefit from that
growth. Recent World Bank reports show that the best policies
for poverty reduction involve more redistribution of influence,
advantage and subsidies away from wealthier, more powerful
groups to those that are disadvantaged.11 Countries that have
attempted to address inequality by investing in the health,
housing and education of their most vulnerable populations
tend to perform better on all human development indicators,
including GDP.12 Countries such as Brazil, Cuba, Egypt,
South Africa, Sri Lanka, Thailand and Tunisia, for instance,
have performed relatively well on many human development
indicators and have managed to contain or reduce slum
growth because of a political commitment – backed by
resources – to invest in the urban poor. Inclusive and vision-
ary urban planning and governance that includes slum
upgrading and prevention, combined with pro-poor urban
development policies that expand and improve opportunities
for employment are, therefore, key ingredients for sustainable
urban development; these are also key ingredients for the
achievement of the Millennium Development Goals in cities.
49
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

THE BAD NEWS
The locus of poverty is shifting to cities
Despite the enormous potential of cities to reduce poverty
and to achieve the Millennium Development Goals, recent evi-
dence shows that the wealth generated by cities does not auto-
matically lead to poverty reduction; on the contrary, intra-city
inequalities are on the rise, particularly in the cities of Africa
and Latin America.
In fact, urbanization in many developing countries, particu-
larly in sub-Saharan Africa, has not been accompanied by eco-
nomic growth, industrialization or even by development per
se.13 On the contrary, the population of some African cities has
grown despite poor economic growth; the region as a whole has
the highest urban growth rate in the world, at 4.58 per cent per
year. This phenomenon, combined with inequitable distribu-
tion of resources and anti-poor policies, has led to rising urban
poverty, which impedes the sustainability of cities and impacts
their economic viability. In many parts of the region, high rates
of urban population growth, high prevalence of unskilled labour
and the HIV/AIDS pandemic are further undermining poverty
reduction efforts in cities. Even in Asia’s economically successful
and rapidly industrializing countries, such as China and India,
urban poverty remains a persistent problem as national GDP
rates have risen much more quickly than national poverty rates
have fallen.14 The economic growth models used by govern-
ments and local authorities have widened not only disparities
between rural and urban populations, but also inequalities
between high- and low-income populations within cities.
Poverty is already becoming a severe, pervasive and largely
unacknowledged feature of urban life. Poverty is shifting to
urban areas and growing in magnitude. World Bank estimates
indicate that while rural areas are currently home to a majority
of the world’s poor, by 2035, cities will become the predomi-
nant sites of poverty.15 But in Africa, the proportion of people
living in poverty in urban areas (43 per cent) is catching up
much faster with the proportion of people living in poverty in
rural areas (59 per cent).16 Sub-Saharan African countries have
some of the world’s highest levels of urban poverty, extending to
more than 50 per cent of the urban population in the poorest
countries, including Chad, Niger and Sierra Leone. In other
countries – notably Nigeria – urban and rural poverty percent-
ages are almost equal.17 In Latin America, the most urbanized
region in the developing world, there are more poor people liv-
ing in cities than in rural areas. In 1999, only 77 million of the
region’s 211 million poor lived in rural areas, while the remain-
ing 134 million lived in urban areas. Proportionally, however,
far more of those living in rural areas than in urban areas were
poor: 64 per cent of the rural population lived in poverty, as
opposed to 34 per cent of the urban population; levels of depri-
vation are also more extreme in rural areas than in urban areas.18
The picture is quite different in the Caribbean countries, where
urban poverty levels already exceed rural poverty levels.19
Relatively low levels of urban poverty exist in countries of
Northern Africa and Western Asia, where urban poverty levels
are near or below 20 per cent; the highest prevalence of urban
poverty in Asia is in India, at 30 per cent.
UN-HABITAT analyses have further shown that people liv-
ing in slums – where a large proportion, but not all, of the
urban poor live – have worse health outcomes and are more
likely to be affected by child mortality and acute respiratory ill-
nesses than their non-slum counterparts. They are also more
likely to live in or near hazardous locations with few basic serv-
ices, making them more vulnerable to natural disasters such as
floods, and saddling them with heavy health and social bur-
dens, which ultimately affect their productivity.
Despite the existence of increasingly large pockets of depri-
vation within cities, many governments continue to assume
that poverty is mainly a rural phenomenon and that those who
live in or move to cities escape the worst consequences of this
scourge, including hunger, illiteracy and disease. A prevalent
view among governments and the international development
community is that urban poverty is a transient phenomenon
of rural-to-urban migration and will disappear as cities devel-
op, thus absorbing the poor into the mainstream of urban
society. This view is reflected in most national poverty reduc-
tion strategies, which remain rural-focused, and in interna-
tional donor assistance to cities, which continues to be modest
in scale and impact,20 with the result that both national and
international interventions during the last two decades have
had the net effect of increasing poverty, exclusion and inequal-
ity in cities.21
The concept of cities as islands of privilege and opportunity
is supported by national and international statistics on health,
education and income, which generally reflect better outcomes
in urban areas. What these statistics fail to reveal are the severe
inequalities within cities, and the various dimensions of urban
poverty that are not captured by income-based indicators,
including political exclusion and poor quality, hazardous and
insecure housing.

FIGURE 1.4.2 THE ANNUAL GROWTH RATE OF CITIES AND SLUMS

Percentage (%)
2
Slums in Port of Spain, Trinidad NOEL P NORTON/UNEP/STILL PICTURES

-1
be ca

As rn

ric n

As rn

As rn

As rn

ric n

w ped
Af her

Af ara
he

te

te
rib eri
an

ia

ia

ia

ia

ld
st

lo
as

es
ut

rt

or
Ca m

Ea

ve
Sa
No

W
E
So
e A

h-

De
b-
th atin

ut

Su
So
an
d L

Urban growth rate Slum growth rate

Source: UN-HABITAT, Global Urban Observatory 2004.

50 51
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Poverty is becoming a severe,
pervasive and largely unacknowledged
feature of urban life.
The high cost of non-food items, such as transport, health,
education and water in cities, coupled with poor living condi-
tions, including inadequate housing and poor access to basic
services, impact the ability of the urban poor to rise out of
poverty. If the definition of poverty is broadened to include the
social assets available to the poor and their vulnerability to
stress and shocks, including evictions, crime, disease, environ-
mental disasters and unpredictable employment markets, then
it is likely that the proportion of poor people living in cities is
much higher than current estimates. Poverty reduction inter-
ventions, including efforts to achieve the Millennium
Development Goals, therefore, have to be focused on cities of
the developing world, which are absorbing a significant num-
ber of the world’s poor and where slums are growing at
unprecedented rates.22
By 2007, half the world’s population will be urban, and in the
next two decades, more than 95 per cent of the population
growth in the world’s poorest regions will occur in cities. Urban
growth rates are particularly high in the least developed countries,
at almost 5 per cent per year. This shift in population implies that
the major development challenges – and the struggle to achieve
the Millennium Development Goals and targets – will have to be
focused on cities of the developing world, where an increasingly
large proportion of the world’s poor will live.
■ Achieving the slum target
Millennium Development Goal 7, target 11 on improving
the lives of at least 100 million slum dwellers by 2020 forces
the international community to address one specific aspect of
urban poverty that until now was neither captured in national
statistics nor reflected in urban data. The slum target is a
recognition by the international community that slums cannot
be considered an unfortunate by-product of urbanization;
rather, by ignoring the plight of slum dwellers, governments
are inadvertently adopting urbanization models that are nei-
ther sustainable nor acceptable.
Slums currently house one out of every three urban dwellers.
Although slums23 represent a physical manifestation of urban
poverty and do not capture the myriad facets of urban pover-
ty that have little to do with housing or basic services, their
rapid growth in the last 50 years, particularly in Asia and
Africa, indicates that they are home to a large number of the
world’s urban poor who are not benefiting from the wealth
52
and opportunities generated by the cities in which they live.
Slums are becoming the norm rather than the exception in Endnotes
the poorest cities of the world. In sub-Saharan Africa, where
more than 70 per cent of urban residents live in slums, many 1 The campaign to eradicate poverty began in earnest in 2000 when
world leaders pledged to halve it by 2015 at the UN Millennium
slum dwellers are unable to escape the material deprivation,
Summit, held in New York in September 2000.
illiteracy and disease that are normally associated with impov-
erished rural areas. Asia’s cities, which host almost 60 per cent 2 The Millennium Development Goals and targets are derived from the
of the world’s slum population, are becoming sites of severe Millennium Declaration adopted by the world’s governments at the
environmental degradation and pollution, which are impact- Millennium Summit in 2000.
ing recent economic gains. And despite progressive legislation 3 World Bank 2000.
and improved governance structures in recent years, Latin 4 GDP refers to the total value of goods and services produced within a
Fruit vendor in Shanghai, China ©AMARK STEPHAN. IMAGE FROM BIGSTOCKPHOTO.COM
America’s cities remain the most unequal in the world. country by both nationals of the country and foreigners.
The sheer scale of the problem warrants attention. Nearly 5 GNP refers to the total value of goods and services produced by nation-
one billion people around the world are currently living in als of the country.
urban slums that lack basic services and adequate housing, and 6 World Bank 2000.
their numbers will increase if no remedial action is taken. 7 U.S. Conference of Mayors 2004.
While the target of improving the lives of at least 100 million 8 Mehta 2004.
slum dwellers may be achieved by 2020, particularly in coun- 9 UN-HABITAT/DFID 2002.
tries that have put in place slum upgrading and prevention 10 Hardoy, et al. 2001.
policies, the scale of the problem may worsen: UN-HABITAT 11 This point has been extensively argued in the World Bank’s World
estimates indicate that if governments continue with business Development Report 2006, which shows that the best policies for
as usual, then an additional 400 million people will be drawn poverty reduction involve more redistribution of influence, advantage or
into the misery of slum life as the global slum population subsidies away from dominant groups.
reaches 1.4 billion in 2020. 12 This finding, for instance, is reflected in UNDP’s Human Development
However, slums do not simply ensnare impoverished urban Report 2005.
dwellers; they also act as intermediate urban spaces, situated 13 Davis 2004.
between destitution and opportunity – key places of transition 14 This fact was highlighted at the 2005 Asia-Pacific Economic
that can help or hurt individuals, depending on the actions of Cooperation Summit in Busan.
governments, the private sector, civil society and slum dwellers 15 Ravallion 2001.
themselves. They can also provide upward mobility to urban Development Goals and targets. Improving the living condi- 16 United Nations 2005a.
dwellers and become sites of immense economic opportunity, tions of slum dwellers, by improving housing, tenure security 17 World Bank 2002 estimates.
culture and innovation – the hallmarks of successful cities. and access to water and sanitation, will automatically have a 18 United Nations 2005a.
Despite the substandard living conditions prevalent within positive impact on the attainment of most of the Goals and 19 Inter-American Development Bank 2004.
them, slums can also represent a kind of opportunity for the targets. In regions where slum dwellers do not suffer from 20 Cohen 2004. Author estimates that total urban assistance to develop-
urban poor. In most regions, slums are “stepping stones” out multiple shelter deprivations, interventions and investments in ing countries from 1970 to 2000 was just $2 billion a year.
of rural poverty, as slum dwellers have more urban-based just one sector can dramatically reduce the numbers of people 21 UN-HABITAT 2003a.
employment opportunities than villagers and have better living under slum conditions. 22 For a detailed analysis of this phenomenon, see Chapter 1.2.
access to publicly financed services and infrastructure. Indeed, UN-HABITAT is convinced that the slum target cannot be 23 See Chapter 1.2 for the UN-HABITAT definition of “slums”.
most slum dwellers choose to remain in cities because of the achieved in isolation. On the contrary, it is becoming increas- 24 Babar Mumtaz makes a case for why “Cities Need Slums” in Habitat
perceived and actual social and economic benefits they pro- ingly clear that the failure of the slum target will jeopardize the Debate, Vol 7. No. 3, September 2001. He argues that insisting on a
vide. Slums, some have argued, are an integral and natural part achievement of all the other Goals and targets; conversely, “city without slums”, especially when no alternative housing has been
of economic growth and industrialization and should be con- achieving the other Goals and targets in slums will make the developed, can mean even more hardship for the very group that is
sidered marks of “success” in urban areas. They reflect a achievement of the slum target more likely.25 essential to urban development: the rural migrant.
dynamic and diverse labour market and offer affordable hous- The struggle to achieve the Millennium Development Goals 25 Moreno 2005.
ing to those who cannot, or will not, pay more for accommo- has to be waged in slums, not at the expense of rural areas, but
dation in the city.24 alongside them. Ultimately, as the world becomes more urban,
Cities and the slums within them also offer governments an the battle to achieve the Goals will be won – or lost – in the
“opportunity” or entry point to tackle some of the world’s zopadpattis of Mumbai, the bidonvilles of Abidjan, the chawls
most pressing challenges, including extreme poverty, under- of Ahmedabad, the villas miseria of Buenos Aires, the favelas of
five mortality, HIV/AIDS, environmental degradation, and Rio de Janeiro, the barrios ilegales of Quito, the shammasas of
gender inequality. The sheer concentration of people living in Khartoum, the iskwaters of Manila, the chereka betes of Addis
cities and slums means that any investment is likely to reap Ababa, the aashwai’is of Cairo, the corticos of São Paulo, the
greater benefits per capita. On the other hand, the economies colonias populares of Mexico City, the vijiji of Nairobi, the
of scale offered by high density slums also make them ideal tar- gecekondus of Ankara, the hoods of Los Angeles, the museques
gets for interventions aimed at achieving the Millennium of Luanda and the katchi abadis of Karachi.
53
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6

Cities, Slums and the Millennium Development Goals Goal 4: Reduce under-five mortality
Goal 1: Eradicate extreme poverty and hunger Reduce by two-thirds, between 1990 and 2015, the under-five mortality rate

Halve, between 1990 and 2015, the proportion of people whose income is less than $1 a day  Better access to health facilities in
Under-five mortality rates are higher in slums than in non-slum urban areas.
Halve, between 1990 and 2015, the proportion of people who suffer from hunger. cities means that children born in High child mortality rates in slums are not so much related to whether or not
urban areas have a better chance of children are immunized; rather, they have more to do environmental factors,

 Cities act as catalysts for poverty reduction: they generate the wealth
The locus of poverty is moving to cities. In the surviving than their rural counter-
parts, who may not live near facili-
such as overcrowding, indoor air pollution, poor wastewater treatment and
lack of drainage, sewerage and sanitation facilities. The use of solid fuels,

A victim of eviction in Zimbabwe UN-HABITAT

and the economic opportunities needed to make the achievement of next two decades, more than 95 per cent of the ties that offer services such as combined with overcrowding and poor ventilation, in slum households increas-
the Millennium Development Goals possible. population growth in the world’s poorest regions immunization and post-natal care. es the chances of children contracting acute respiratory illnesses, such as
will occur in urban areas, with the result that
 Urbanization levels are closely related to levels of income and better cities will become the predominant sites of  Empirical evidence suggests that
pneumonia. Many slums are also located in or near hazardous or toxic sites,
which expose children to additional environmental and health hazards.
performance on social indicators, including health and literacy. The poverty in coming years. sectoral interventions in water, san-
achievement of the Millennium Development Goals is, therefore, more
Access to more health care facilities in urban areas does not automatically
likely in cities. Conversely, urban economic growth provides the basis
Malnutrition, hunger and disease are becoming itation and housing have positive
outcomes in the reduction of child lead to reduced mortality rates in slums. Parents struggling to pay for food,
on which cities can contribute to the achievement of the Goals, partic- more prevalent in slums, particularly in develop- mortality rates. school fees and transport costs may be unwilling or unable to pay for the
ularly in the area of poverty reduction. ing countries. Because hunger experienced in health care of their children, which has an impact on child mortality rates.
Somalia UNICEF

cities is directly related to income (rather than

 Cities and slums are often the “first step” out of rural poverty. The agricultural productivity), the urban poor are
rural poor move to cities, where there are more employment opportu- much more vulnerable to income-dependant
nities and better access to services such as health care and education. hunger than their rural counterparts. Goal 5: Improve maternal health
Reduce by three-quarters, between 1990 and 2015, the maternal mortality ratio
Goal 2: Achieve universal primary education
 Mothers are more likely to be
Many women living in slums cannot afford the relatively more expensive
Ensure that, by 2015, children everywhere, boys and girls alike, will be able to complete a full course of primary schooling attended by pre-natal healthcare delivery and post-natal health services in urban areas, and are, therefore,
personnel in cities, as there are less likely to seek these services, especially when a choice has to be made
 Cities are nodes of edu-
Educational facilities are generally more advanced and accessible in cities

Luanda, Angola EDUARDO LÓPEZ MORENO

cation and learning, a key
contributing factor in
Education in Africa UNEP/STILL PICTURES
generally more health care facil- between paying for these services and buying food or meeting other
ities in urban areas than in rural household expenses.
but the cost of other items (transport, housing, food) is higher in cities than
areas. Increased awareness in
in rural settings, which impacts the ability of the poorest households to

Poor urban women who supplement their incomes by engaging in sexually

urban areas means that women

rural-to-urban migration. send children to school. Slum children, particularly girls, are more at risk of
are more likely to have a better risky behaviour expose themselves to a variety of sexually transmitted dis-
Urbanization has been dropping out of school than children living in non-slum urban areas.
understanding of the need for eases that have a negative impact on maternal health. Numerous studies
associated with econom-
ic and social progress,
In many slums, overcrowding and inadequate or non-existent toilet facilities ante-natal and post-natal care
than their rural counterparts.
have shown high rates of HIV infection among urban women; in sub-
Saharan Africa, HIV prevalence is highest among women living in urban
the promotion of literacy and other amenities further impact the quality of education that children
areas, and particularly high among women living in slums.
and education. receive. In some countries, adolescent girls drop out of school because of
insufficient toilet facilities in slum schools, or because of family responsi-
bilities, such as taking care of siblings while parents are at work.
Goal 6: Combat diseases including HIV/AIDS, malaria and other diseases
Goal 3: Promote gender equality and empower women Have halted by 2015 and begun to reverse the spread of HIV/AIDS
Have halted by 2015 and begun to reverse the incidence of malaria and other major diseases
Eliminate gender disparity in primary and secondary education, preferably by 2005, and in all levels of education no later than 2015
 Access to information and
Increased awareness about prevention in urban areas has not had the desired effect of reducing HIV prevalence in
 Cities offer women social mobil-
With the exception of Africa, the share of woman-headed households is greater in urban areas than in rural areas. health care facilities in cities is cities. In fact, trends suggest that HIV prevalence is much greater in urban areas than in rural areas, and is also high-
ity, which has a positive impact This trend has socio-economic implications, which can translate into deepening poverty among urban women in higher than in rural areas. er among urban women than among rural women.
on gender equality and has
helped reverse some socially
some countries. Increased awareness has led to
the prevention of malaria and
The HIV-AIDS pandemic may shave off up to 2 per cent of annual economic growth in the worst affected countries. This
prescribed roles. In many coun-
In situations of limited resources and urban impoverishment, women and girls are the first to be sacrificed when other major diseases and has a direct impact on urban economic growth and in some countries severely undermines poverty alleviation efforts.

tries, urban women have more
access to land and property than
their rural counterparts as they
are not constrained by discrimi-
natory customary laws. In gener-
al, urbanization has had a posi-
tive impact on women’s access
to resources and enlarged their
decision-making roles.
families have to make difficult choices about sending their children to school. widened people’s choices in the
The situation of extreme deprivation in cities, particularly in slums, encourages residents to engage in risky sexual

Slum life forces many women and girls to engage in sexually risky behaviour, making them more vulnerable to
treatment and prevention of the behaviour for economic survival. Slum residents often start sexual intercourse younger, have more sexual partners,
HIV/AIDS epidemic. and are less likely than other city residents to know of or adopt preventive measures against contracting HIV/AIDS.
HIV/AIDS and other sexually transmitted diseases. HIV prevalence among urban women in sub-Saharan Africa is
 Slum upgrading and prevention
HIV/AIDS has contributed to the growing problems of AIDS orphans; many of these orphans become street children
already is much higher than among rural women.

Poor access to water and sanitation places an enormous labour and health burden on women living in slums, who are policies that incorporate volun- caught in the poverty trap of hunger, malnutrition, disease and illiteracy.
not only charged with ensuring that their families have water, but who also suffer disproportionately from the health
tary testing and counselling
facilities have helped reduce HIV

Slums are characterized by overcrowding and poor ventilation, the leading contributors to the rise in tuberculosis
and environmental hazards associated with poor sanitation. In slums, where there are few or no toilets, many women cases worldwide. Studies have shown that HIV-related tuberculosis is becoming an increasingly urban phenomenon,
prevalence in urban areas.
are forced to defecate in the cover of darkness, which renders them more vulnerable to sexual and physical assault. particularly in slums.

54 55
 Goal 7: Ensure Environmental sustainability Goal 8: Develop a global partnership for development
Integrate the principles of sustainable development into country policies and programmes and reverse the loss of environmental resources Address the special needs of the least developed countries and small island developing States
Halve, by 2015, the proportion of people without access to safe drinking water and basic sanitation. Develop further an open, rule-based, predicable, non-discriminatory trading and financial system
By 2020, to have achieved a significant improvement in the lives of at least 100 million slum dwellers (Goal 7, Target 11) Deal comprehensively with developing countries’ debt
In cooperation with developing countries, develop and implement strategies for decent and productive work for youth
In cooperation with pharmaceutical companies, provide access to affordable and essential drug in developing countries
In cooperation with the private sector, make available the benefits of new technologies, especially information and communications

Sustainable urban development

Air Pollution ©ALISTAIR SCOTT. IMAGE FROM BIGSTOCKPHOTO.COM

Urbanization can bring about irreversible changes in production

and consumption of water, energy and land. Both developed and
developing countries are witnessing rapid urban sprawl with
direct consequences for the surrounding hinterland.

Air pollution is concentrated in cities. The concentration of

Sustainable urban development
 Sustainable urbanization policies have been incorporat-
ed in many city and national plans and are contributing
to reversing the impact of environmental degradation
and pollution. Cities with sound and sustainable land,
air and water management policies have also managed
industrial emissions and increased motorized transport in cities
is severely eroding their environmental sustainability and is
affecting the health of urban populations. Acute respiratory ill-
nesses associated with poor air quality and poor housing condi-
tions are impacting the human and economic productivity of
cities, particularly in Asia; it is estimated that the health costs
from pollution reduce gross domestic product (GDP) by some 2
per cent in developing countries.
Slums

The rate at which slums are growing exceeds the rate at which
they are being improved. This severely impacts the achievement
of Goal 7, target 11: by 2020, to have achieved a significant
improvement in the lives of at least 100 million slum dwellers.

Tokyo NATSUO ITO

to reduce soil erosion, improve air and water quality
and protect biodiversity within cities and in their hin-
terlands. Some cities in the developed world are uni-
laterally reducing greenhouse gas emissions and other
pollutants as part of their respective governments’
commitment to adhere to the Kyoto Protocol.
 Cities concentrate production and population, which
gives them obvious advantages over rural settlements
or dispersed populations. For example, the concentra-
tion of populations in urban areas greatly reduces the
unit cost of piped water, sewers, drains and roads. The
use of environmentally friendly energy sources and
transport can reduce these costs even further.
Slums
 Slums provide an important entry point for the achieve-
ment of all the Millennium Development Goals in
cities; the sheer concentration of people living in slums
make them ideal targets for interventions aimed at
reducing poverty, reducing child mortality and
HIV/AIDS, improving literacy and promoting environ-
mental sustainability in urban areas.
Some 200 million more slum dwellers have been added to the
world’s urban population since 2000; if current trends continue,
by 2020, there will be 400 million more people drawn into the
misery of slum life and the global slum population will reach
1.4 billion.

Slum dwellers are more likely to live in hazardous or toxic loca-
tions, which are more prone to natural disasters, such as
floods, and which pose severe health risks, not just to slum
dwellers but to city dwellers in general.

Although access to water and sanitation is generally better in
urban areas than in rural areas globally, the consequences of
poor access in cities are more severe. Many slum dwellers have
no choice but to use water sources, such as rivers, to bathe and
wash clothes. Poor sanitation in some cities has also led to
large sections of the population defecating in the open. This
contributes to contamination of water and land resources with-
in cities, and is a cause of many of the water-borne diseases
prevalent in slums.

Indoor air pollution caused by the use of solid fuels is prevalent
in slums and is a leading cause of respiratory illnesses in urban
areas, particularly among women and children.
Partnerships for development
 In the last decade, the international community has become much more aware of the
problems, challenges and opportunities of urbanization. City-to-city cooperation and
decentralization policies are becoming more common as governments at the local and
national levels cooperate to make cities more liveable. Some national governments are
for the first time addressing urban poverty in their national poverty reduction strategies
and programmes. Many local governments are also recognizing the need to plan, manage
and govern their cities better, which has improved the lives of many urban residents
worldwide. As part of new structures of governance, where cities are being given more
authority to manage their affairs, cities are playing a more prominent role in developing
partnerships with central governments, regional organizations and development partners
to promote sustainable urbanization within cities, countries and regions.
 Some multilateral agencies and regional development banks are recognizing the need to
intervene and invest in urban areas as part of their development assistance programmes.
In Latin America, for instance, the portfolio of loans for urban development have grown
in both volume and complexity.
Partnerships for development

Increased awareness of the potential and chal-
lenges of cities has not led to a commensurate
increase in international development assistance
or to greater allocation of national finances to
urban poverty reduction. Most development aid is
focused on eradicating extreme poverty and
improving the living conditions of rural popula-
tions. Many governments have not identified
“urban poverty” as an area of intervention in their
development plans.

Although investments have been made in various
sectors, such as health and infrastructure develop-
ment, they are often not tagetted specifically at
slums.

56 57
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

2
Part Two

The State of the

World’s Slums

Slums are the emerging human settlements of the

21st century. In order to monitor the state of the
world’s slums, this Part presents a detailed
analysis of the five indicators that reflect
conditions that characterize slums. These
indicators, known as “shelter deprivations”, are:
lack of durable housing; lack of sufficient living
area; lack of access to improved water; lack of
access to improved sanitation; and lack of secure
tenure. Information on the five indicators is
analysed at global, regional, national and city
levels. This Part also presents an analysis of the
degrees of shelter deprivation in some selected
countries and regions.
Manila TINA ELAINE RUSTE/UNEP/STILL PICTURES
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
2.1 Neither Brick nor Mortar: Non-Durable Housing in Cities
D
urability is one of the least understood attrib-
utes of a nation’s housing stock, and life
expectancy of a house is a neglected indica-
tor.1 In many parts of the world, housing
durability assessments are not systematic, and
when conducted, they do not generate data that is comparable
with data in other areas. Estimates suggest that worldwide, 18
per cent of all urban housing units (some 125 million units) are
non-permanent structures, and 25 per cent (175 million units)
do not conform to urban building codes or regulations.2 These
figures, however, could be highly underestimated as global data
on durability is based primarily on permanence of individual
structures, not on location or compliance with building codes.
Mainstream reporting mechanisms are not designed to capture
data on unsafe or hazardous location of housing, but com-
pelling reasons exist for collecting such data, as it is estimated
that at least three or four in every 10 non-permanent houses in
cities in developing countries are located in dangerous areas
that are prone to floods, landslides and other natural disasters.3
Other unsafe locations include living on garbage dumpsites or
in highly polluted areas. Non-durable dwellings located on haz-
ardous locations are particularly at risk when natural disasters
strike, as they are least able to withstand the destruction caused
by flood, earthquakes or hurricanes. However, people living in
extreme poverty in cities often have little choice but to take
shelter in ramshackle structures haphazardly constructed in the
most polluted, dangerous areas.
Informal settlements in the developing world typically do not
meet local building codes and other regulations for urban devel-
opment. Disregard of building codes has clear implications for
the durability and safety of housing. The destruction of residen-
tial buildings wrought by recent earthquakes in cities – such as
those in Turkey in 1999, India in 2001 and Iran in 2003 – con-
firmed the importance of enforcing building codes and regula-
tions, as the structures that collapsed were found to have violat-
ed building norms and technical requirements deemed neces-
sary in earthquake-prone regions. Poor enforcement of building
codes was also the primary reason cited for the collapse of a
multi-storey building under construction in Nairobi, Kenya, in
January 2006, which claimed 14 lives and injured several oth-
ers.4 The urban poor suffer the greatest loss when natural disas-
ters strike, in part because their housing is built to lower stan-
dards than housing for higher-income residents.
In the developed regions, it is estimated that almost all hous-
ing is in compliance with codes and regulations. The Canadian
government, for instance, reports that just a fraction of the
country’s housing stock – 5 to 10 per cent – is non-compliant,
60
It is estimated that at least three or four
in every 10 non-permanent houses in
Buenos Aires MARK EDWARDS/STILL PICTURES
cities in developing countries are
located in dangerous areas that are
prone to floods, landslides and other
natural disasters.
comprising older construction that has not been upgraded or
add-on structures that have not been brought up to the stan-
dard of the rest of the dwelling.5 In some European cities, a
small amount of unauthorized construction of dwelling units
takes place in absence of building permits and in violation of
zoning rules. Such houses are illegal in the technical sense, but
in most cases, they respect current building standards. There
are, however, a small number of houses that do not comply
with building codes, some of which are regarded as unhealthy
dwellings. In Naples, such buildings are called “bassos”; in
Barcelona, “illegal pensions” are a problem; and in Paris, many
low-income families live in illegal boarding houses and other
forms of sub-divided apartments.6
In Europe, North America and other developed regions,
housing durability has undergone a fair amount of analysis,
particularly with regard to building materials, maintenance and
construction methods and systems. Issues such as affordability,
accessibility, financial costs, and quality of the dwelling are rel-
atively well researched. Studies also include measures to
improve housing durability, not only for economic reasons, but
also for disaster mitigation and vulnerability reduction.7
The vast majority of housing in developed countries is in
decent condition. In Canada in 2001, for example, just one in
12 homes (8.2 per cent) were in need of major repairs, and
nearly two-thirds of homes needed only regular maintenance.8
In the United States, approximately 2 per cent of occupied
units had severe physical problems with plumbing, heating,
electricity, public areas, or maintenance in 1999.9 This ratio is
similar to other developed countries that have a housing stock
comprised almost entirely of permanent buildings, even in the
lowest-income parts of the city. In some European cities, a few
unique exceptions exist, such as the Roma encampments built
with temporary or “non-permanent” materials.
An interesting pattern emerges from the data about the hous-
ing stock in developed countries, however, which reveals that
the units that are in the worst condition and require the most
repairs often house members of ethnic minorities or immi-
grants.10 Many such houses are overcrowded and dilapidated
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

MAP 7 PROPORTION OF HOUSEHOLDS WITHOUT FINISHED MAIN FLOOR MATERIAL, 2003

Affordability, Accessibility and Durable Houses
Durability – the longevity of residential struc-
tures – is directly associated with housing
accessibility and affordability.
In the formal housing market in developing
countries, mortgage institutions, such as com-
mercial or trading banks, restrict loans to
builders who use materials that are “accept-
able” in the market because they are considered
“durable” (e.g. bricks, stones, concrete, and the
like). The use of durable materials decreases
the bank’s risk over the period of the mortgage
loan. This housing policy excludes local building
materials such as wood, bamboo or other inno-
vations that arise out of the use of indigenous
materials; such materials are inexpensive, but
are subject to rapid deterioration and subse-
quent depreciation. Financial institutions, there-
fore, have adhered to standards and building
codes in a way that limits access to affordability
schemes for the working poor.
In the informal housing sector, slum dwellers
recognize the trade-offs between durability and
affordability. They understand that a durable
house has a clear positive economic benefit, is
safer and can be healthier for its occupants than
one made of temporary materials. However, put-
ting up a house that lasts requires a high initial
investment, even though it could cost less in the
long run. “Building to last” in terms of using
stronger and more durable materials that would
reduce maintenance and replacement costs
remains out of the financial reach of most slum
dwellers. Poor families requiring housing right
away are more likely to opt for temporary solu-
tions than wait until they can afford something
more permanent. These solutions increase their
vulnerability, as the units may be structurally
unstable or located in hazardous areas.
Lack of affordable land and the absence of
affordable self-build housing schemes leave
impoverished urban residents susceptible to
risks and harms that could be alleviated by
improved urban planning and the development
of more affordable building materials that are >45%
durable and easy to use by self-builders. >25-45%
>10-25%
<=10%

dwellings located in low-income neighbourhoods, built in the
1970s or earlier as part of government housing development
projects, in older parts of central cities and first-ring suburbs.
Many are substandard units that are contaminated and lacking
light, air and open space. Others are poorly built, poorly main-
tained and isolated, often situated in inaccessible or unhealthy
locations, such as along motorways and industrial wastelands.
Even if the number of these housing units is statistically
insignificant, they typify some of the exclusion patterns and
forms of physical and social decay prevalent among impover-
ished minorities in the developed world. Government agencies
have also observed that rental units often have twice as many
durability problems as owner-occupied units.11
A close relationship clearly exists among durability, afford-
ability and accessibility in housing markets and building pat-
terns around the world. These linkages are very often neglected
and thus are not properly analyzed for policy purposes.
■ UN-HABITAT data and analysis of housing
durability in developing countries
To estimate the prevalence of slums around the world using data
collected between 1990 and 2001, UN-HABITAT included a
London MJS
Source: UN-HABITAT 2005, Global Urban Observatory, Urban Indicators Programme, Phase III.
tial problems with other key structural materials of dwellings. fore needed to monitor spatial inequalities within cities where
For example, if housing durability estimations include quality segregated urban social structures persist.16 UN-HABITAT’s
of roof and wall materials, the figure for many countries would urban inequities study conducted in Addis Ababa, Ethiopia, for
drop. (See table.) For instance, in Bolivia, when only floor instance, highlights the disparity between slum dwellers and
material is considered, 83.8 per cent of the urban population is other urban groups with regard to adequate housing. While
counted as living in durable housing, but when wall and roof more than half (51.6 per cent) of slum households used natu-
materials are considered, this figure drops to 27.7 per cent. In ral, non-permanent, materials for the floors of their dwellings,
Nicaragua, when floor and roof materials are combined to 58.4 per cent of non-slum households used polished cement, a
determine durability, only 9 per cent of the urban population permanent material. Also, a significantly higher percentage of
qualifies as living in a durable home.
In order to produce more accurate data on housing durabili-
ty, a statistical analysis was conducted in the countries where TABLE 2.1.1 HOUSING DURABILITY, BASED ON FLOOR, ROOF AND
information is available for the three main physical structure WALL MATERIALS, IN THE URBAN AREAS OF 16 SELECTED
variables – floor, walls and roof – at the urban level. The results COUNTRIES, 2001
provide a more realistic image of housing durability, and three
Country Floor Wall Roof All
examples aptly illustrate this point. In Indonesia, the percent-
age of durable housing in urban areas in 2002 was 69.8 per cent
Benin 80.2 61.6 88.3 60.1
when the three components of the dwelling were considered,
whereas when only the floor criterion was used, 83.7 per cent Central African
Republic 26.2 9.1 52.5 7.7
of houses were deemed durable.12 In Benin, 80 per cent of
Chad 15.3 5.3 52.9 4.7
houses qualified as permanent and durable in 2001, consider-
Togo 94.4 72.6 88.7 66.2

measurement of housing durability as one of its five indicators of ing only the floor criterion; however, when materials for the
Uganda 68.6 58.5 91.8 53.7
slum households. In principle, the estimation procedure consid- three elements of the house were taken into account, housing
durability dropped to 60 per cent.15 UN-HABITAT results are Bolivia 83.8 52.0 41.5 27.7
ered the nature of the roof, wall and floor materials of dwellings.
Data on all three was easily obtainable for developed countries, but consistent with data produced by governmental sources that Brazil 89.2 95.2 98.9 86.1
in the developing world, estimations were made considering only assess quality of housing combining the three variables. India, Dominican Republic 95.7 92.5 98.6 88.6
the nature of the floor material. Roof and wall materials were con- for instance, reported that 73 per cent of urban households Guatemala 77.4 66.3 96.7 60.8
sidered inappropriate variables for the durability indicator, as infor- lived in pucca or permanent houses in 1991,13 and Sri Lanka Nicaragua 65.3 61.2 21.9 9.0
mation on them is collected in few countries. reported that permanent structures with brick walls, tiled roofs, Peru 66.5 58.4 84.8 47.7
The research revealed that in 2003, 94 per cent of the world’s and cement floors constituted 70 per cent of houses in urban Bangladesh 53.2 64.3 27.3 26.4
housing units in urban areas were considered permanent based on areas in the early 1990s.14 Indonesia 83.7 72.6 93.9 69.8
the “floor criterion”, meaning that most of the floor in each This data only presents the national and urban aggregates, 69.2 59.2 72.1 46.8
dwelling was constructed or covered with permanent materials and which are useful for monitoring urban poverty and sustainabil-
was not simply earthen. Using this criterion alone masks poten- ity at city, national and regional levels. Further research is there- Source: UN-HABITAT, Global Urban Observatory, 2005.

62 63
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

slum dwellers (74.6 per cent) used traditional (non-permanent)
materials to construct their walls, compared to non-slum
dwellers (58.7 per cent). While UN-HABITAT recognizes that
slums are not always geographically contiguous, in most cities,
slum households are clustered to some extent, indicating specif-
increasing housing durability has been registered in the interme-
diate cities of Assyut, Aswan and Beni Suef in Egypt. These cities
have also made progress on other shelter indicators, leading to an
overall decline in the number of slum households. The city of
Porto Novo, Benin – an important political capital and host of
Defining durability
Presently, global data on housing durability is
not possible to collect and analyze because

Old Town in Fez, Morocco. NATSUO ITO

ic areas of cities in which housing conditions are distinctly an annual international festival – has also shown significant researchers and governments have not agreed
worse than in other areas. It should therefore be emphasized improvements on housing durability and other indicators that upon definitions, specific indicators, classifica-
that an understanding of the spatial patterns of inequality is are attributed to increased government investments in housing. tions, and approaches. What little data has
fundamental to formulating area-based policies to address lack Kigali, the capital city of Rwanda, has experienced a significant been collected has several methodological
of housing durability and other deprivations.17 improvement in housing durability and on all of the slum indi- weaknesses. For instance, wood is considered
cators. Despite the conflict that ravaged the country in 1994, durable in developed regions, but not in most
Rwanda has made steady progress in economic recovery and has developing countries. Other building materials
■ A global and regional overview of housing made impressive sectoral improvements in education and health. are classified as “rudimentary” (e.g. mud or
durability The city of Dar es Salaam, Tanzania, on the other hand, showed palm), but in certain cases, they are recorded as
an improvement in housing durability but did not have similar “permanent”. A material may not be deemed
UN-HABITAT estimates indicate that in 2003, 133 million progress on other shelter indicators. durable in terms of other, more modern building
people living in cities of developing regions lived in housing Urban-level aggregates conceal intra-city disparities. In cities materials, but when combined with skilled con- floors or ceilings, etc.” In Japan, “durable” depending on definitions and measurement cri-
that lacked finished floor materials. In the developing world, such as Luanda, Angola, and Arusha, Tanzania, the population struction and regular repair, it could be coded as housing is any that is not dilapidated, as defined teria. Most of the censuses collected around the
“durable”. Moreover, a problem that arises by the government. According to this definition, world present data on housing quality based on
Asia had the largest proportion (73 per cent) of urban dwellers living in durable houses is approximately 50 per cent.18 It is like-
when measuring the permanency of the struc- housing durability can be captured by three the material used for construction of walls,
living in non-permanent housing. Over 50 per cent of this ly that housing deficits in these countries will increase, as their
ture is that durability manifests itself differently main conditions: permanence of structure, non- roofs and floors separately. The Indian govern-
population lived in Southern Asia, followed by South-Eastern slum growth rate – on average, 5.4 per cent – is higher than the
in different cities. In Nairobi, for instance, non- hazardous location and compliance with local ment, for instance, classifies a house as pucca
Asia (11 per cent). Stark contrasts exist between Northern rest of the sub-region, where the growth of slums is already stag-
durable houses may be made of a patchwork of building codes. when the materials used both for the roof and
Africa and sub-Saharn Africa. Northern Africa hosts only 1 per geringly high, at 4.5 per cent. In other cities such as Port tin, cardboard or plastic sheeting, whereas in wall are regarded as “permanent” (burnt brick,
cent of the developing world’s urban populaton with non- Elizabeth in South Africa, lack of durable housing is surprisingly Mumbai, a temporary house may be made of In developing countries, the nature of the loca- stones, concrete, tiles, and the like). The govern-
durable housing, while sub-Saharan Africa hosts 20 per cent. high if compared with other urban agglomerations in the coun- thatch, bamboo or mud. In Moscow, Tokyo and tion and compliance with building codes are ment classifies houses as kutcha when both
try. Only 58 per cent of Port Elizabeth residents have housing other developed cities, a non-durable unit is rarely used in durability assessments. Whenever materials (roof and wall) correspond to non-per-
Housing durability in Africa considered durable, the lowest in the country. often the equivalent of a dilapidated house a developing country considers the definition of manent (temporary) materials that are replaced
In Northern Africa, housing durability is less of a concern, made of older materials, or an apartment in a durability in a more comprehensive manner, the frequently (grass, bamboo, leaves, mud, and the
Over 10 per cent of the urban population in sub-Saharan since more than 99 per cent of the total urban population lived substandard building. measurement is undertaken by integrating two like). In the cases where there is a mixture of
Africa lives in non-durable housing. Important progress in in durable houses in 2003. variables: “permanence of structure”, under both permanent and temporary materials, the
“Durable housing” is generally defined as a which the state of repair is sometimes also con- dwelling is classified as semi-pucca. (For defini-
“unit that is built on a non-hazardous location sidered; and “hazardous location”, which nor- tions refer to the National Census of India
TABLE 2.1.2 FINISHED FLOOR COVERAGE AMONG URBAN FIGURE 2.1.1 DISTRIBUTION OF THE URBAN POPULATION LACK- and has a structure permanent and adequate mally encompasses compliance with building 2001.) Moreover, very often the settlements’
POPULATION BY REGION, 2003 ING FINISHED FLOOR MATERIALS IN DEVELOPING REGIONS, 2003 enough to protect its inhabitants from the codes. However, no information is provided occupants do not have legal title to their proper-
extreme of climate conditions such as rain, heat, about the description of the area, such as road ty, allowing them to bypass the request for a
cold, and humidity”. conditions and pathways, vehicular access roads building construction permit. Houses built in a
Distribution
Urban of urban pop- and other variables of neighbourhood quality, non-authorized settlement, even if they are per-
population ulation lacking However, the permanence of housing is defined including the type of housing (i.e. detached manent, may be classified as “temporary” until
Access to Urban lacking finished main
Northern Africa by different criteria in different countries. In the houses, multi-storey buildings, and the like). the government grants security of tenure. Since
finished main population finished main floor materials
floor materials, in 2003 floor materials in developing Netherlands, the Department for Housing con- these units are produced by the informal sector,
2003 (%) (thousands) (thousands) world (%) Sub-Saharan Africa siders the number of houses that do not comply People living in poverty in cities endure a com- it is difficult to estimate the number of houses
with the local building codes but are still livable. plex array of housing conditions, the range of built every year and the quality of the structure
Latin America and the Caribbean
In the United States, information about the con- which is difficult to define and capture for sta- and materials in terms of durability.
Northern Africa 98.3 77,910 1,363 1.0 dition of housing, including the need for struc- tistical purposes. Recent migrants may utilize
Eastern Asia
Sub-Saharan Africa 89.1 251,166 27,416 20.6 tural repairs and maintenance, is obtained temporary materials – sometimes scrap – to In developed countries, “compliance with build-
Latin America and Southern Asia through self-reporting in censuses. Similarly, the build their homes and gradually improve them ing codes” is most often used as a separate
the Caribbean 98.2 417,229 7,630 5.7 Canadian government asks its citizens to assess over time. At one stage of this process, the roof variable from the hazardous location of housing.
South-Eastern Asia
Eastern Asia 98.4 564,871 9,271 7.0 the condition of their own dwellings as part of of the house may be made of permanent materi- Thus, all inhabitants of the cities of Amsterdam
Southern Asia 84.8 448,738 68,415 51.4 Western Asia its census every five years. Canadians indicate als, such as galvanized iron sheets or ceramic and Den Haag in the sub-sea level Netherlands
South-Eastern Asia 93.6 228,636 14,650 11.0 whether their dwellings require any major or tiles, while the rest of the unit is made of tem- are classified as living in a disaster-prone area;
Western Asia 96.4 124,370 4,480 3.4 minor repairs, with “major repairs” defined as porary materials. This further complicates classi- yet, their housing units can still be considered
Total 133,226 the need for “repair of defective plumbing or fications, since the same house could be consid- “durable” if the dwellings are in compliance
electrical wiring, structural repairs to walls, ered permanent, semi-permanent or temporary, with the local building codes.

Source: UN-HABITAT (2005), Urban Indicator Programme Phase III and United Nations, World Urbanization Prospects; The 2003 revision. Sources: UN-HABITAT 2003b; Planning Commission of India 2002; Canada Mortgage and Housing Corporation 2005.
Note: Access to finished main floor materials was computed from Demographic and Health Surveys (DHS) data.

64 65
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
AFGHANISTAN RASNA WARAH

Havana ©ALEX BRAMWELL. IMAGE FROM BIGSTOCKPHOTO.COM

Housing durability in Asia
The Asian region is far from homogenous. More than half
inhabitants who lack durable housing in the developing world live
in Southern Asia, a sub-region that has among the highest preva-
lence of slums, infant mortality and poor performance on other
social indicators.19 Around one-third of the urban dwellers in
Bangladesh, Nepal and Pakistan lack durable housing. One-
tenth of the urban population in India, or around 28 million
people, are living in non-permanent structures. In Afghanistan,
despite some advances, housing challenges remain daunting.
Most people in the country live in extreme poverty in cities rav-
aged by war; housing and physical infrastructure in most
Afghan cities need to be rebuilt.20
Apart from Haiti, no reliable data on durability is available for
the Caribbean sub-region. Haiti is notorious for having the worst
social and health indicators in the whole region but surprisingly,
lack of durable housing is not the most prevalent shelter depriva-
tion, as only 10 per cent of the urban population is estimated to
live in non-durable houses. In Haiti, slum dwellers suffer most
from lack of access to improved water and sanitation – less than
50 per cent of the urban population has access to either – and
lack of sufficient living area, or overcrowding, which affects 35 had toilets and 38 per cent had access to an improved water source
FIGURE 2.1.2 PROPORTION OF URBAN HOUSEHOLDS WITH
per cent of the urban population. in 2000, whereas in permanent houses, these ratios were 64 per
FINISHED MAIN FLOOR MATERIALS, BY REGION, 1990 AND 2003
cent and 75 per cent, respectively.22 In addition to being inade-
quate for families’ needs, non-durable housing structures are often
■ Expanding the definition of “durability” situated in urban areas that lack adequate infrastructure, especially
access roads and bridges. When emergencies happen, residents of 100

Throughout Asia, advances in housing durability have been less
conspicuous than in other developing regions, with some second-
ary cities in Indonesia (Bitung and Jaya Pura) experiencing impor-
tant improvements. In 2003, one-fifth of the Asian slum popula-
tion lacked durable housing – a proportion that may be higher,
There is no doubt that housing durability figures for Africa, Asia
and Latin America and the Caribbean are underestimated if the
indicator is not widened to include the condition of floor, wall and
roof materials combined. Methods to measure lack of durability
require further refinement to include more information regarding
such neighbourhoods may find themselves trapped and unable to
escape or call for help. In cases when help does arrive, poor access 80
roads and high densities in the settlement may prevent fire engines
or ambulances from reaching victims. Physical accessibility is con-
60
sidered an essential part of the durable housing indicator, but data

Percentage (%)
considering that these estimates do not include information about compliance with building codes, the hazardous location of residen- collection systems do not as yet capture this vital information,
other construction variables, such as walls and roofs. tial buildings and the condition of individual dwellings. In Brazil, including data on the surroundings of the household unit. 40

Housing durability in Latin America and the
Caribbean
In Latin America and the Caribbean as a whole, the number
for instance, according to information provided by Munic/IBGE
in 2001, all municipalities with more than 500,000 inhabitants
had a certain number of favelas (slums), and most of the favelas had
houses that were non-permanent. Of the municipalities, 87 per
cent had non-authorized subdivisions, called loteamentos clandesti-
As part of the United Nations Expert Group on the 2010 World
Programme on Population and Housing Censuses, UN-HABI- 20
TAT has made recommendations that census questionnaires be
revised and updated to include questions that address three key
0
housing durability variables:

ric n

ric n

be nd

ia

ia

As rn

ia
of non-durable houses in urban areas is relatively low. Durability nos, and 65 per cent showed different forms of inadequate hous- 1. houses in hazardous locations;23

Af her

Af ara

As

As

As
e
rib a a
a

an

ia
st
rt

ah

n
Ea
Ca ric
No

er

er

er
S
ing, including non-durable structures.21

h-
is not a determining factor of slums in the region, since more 2. building codes;24

th Ame

st

th

t
b-

es
ut
Ea

u
Su

W
So

So
tin
e
than 99 per cent of the urban population lived in houses that Studies on housing durability should be expanded to take into 3. and hazard mapping.25

were considered durable in 2003. Overcrowding remains a much
more significant determinant of slum households in the region.
However, serious deficits in housing durability have been report-
ed in Guatemala, Nicaragua and Peru.
La
1990 2003
account other shelter deprivation indicators as well, since non-
durable houses are very often associated with dwellings that lack The next round of censuses will include relevant questions to
some basic services, such as water and sanitation. In Indonesia, for obtain data on these variables to further clarify the extent of hous- Source: UN-HABITAT Urban Indicators Programme Phase III.
instance, of the non-permanent housing stock, only 28 per cent ing durability. Note: Data for 1990 not available for some regions.

66 67
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

Europe’s Forgotten Roma Community

Endnotes
The Roma or Romani (meaning and in France, it is mainly Roma
Roma Settlement in Eastern Europe UN-HABITAT
“man” or “people”) have also asylum seekers from Eastern 1 US Department of Housing and Urban Development 2005. 21 Presidency of the Republic of Brazil 2004.
been called Gypsies, Tsigani, Europe and Asia who end up in 2 UN-HABITAT 2001. 22 Housing Statistics Indonesia 2000.
Tzigane, Cigano, Zigeuner – camps because of official 3 Interviews by Eduardo López Moreno with National Directors of 23 The census will gather additional information on dwellings that are
labels the Roma themselves con- unwillingness to provide proper Housing of various countries as part of the regional workshops on considered hazardous, namely: housing located in areas subject to dis-
sider derogatory. Most Roma housing for refugees. The “Training on Data and Indicators for Monitoring Progress towards the aster more than once every hundred years (disasters include flooding,
identify themselves by their squalor of these camps was Millennium Development Goals.” The countries included, among oth- earthquakes, volcano, storm surge, landslide, or avalanche); housing
tribes or groups, which include revealed in 2004 when two ers, Venezuela, Ethiopia, Indonesia, Angola, and Mexico. not adequately protected against cyclones or bushfires which occur at
the Kalderash, Machavaya, Roma girls from Romania died 4 Ayieko 2006. this frequency; housing settled on garbage mountains or dumps; hous-
Lovari, Churari, Romanichal, in a fire that destroyed the hut 5 In British Columbia, Canada, it is estimated that 20 to 25 per cent of all ing around high industrial-pollution areas; and housing around other
Gitanoes, Kalo, Sinti, Rudari, in which they slept. urban rental housing is in this form, which is 8 to 10 per cent of the total high-risk zones, including railroads, airports and energy transmission
Manush, Boyash, Ungaritza, Luri, housing stock. Canadian Mortgage and Housing Corporation 2005. lines.
Bashaldé, Romungro, and The European Roma 6 Scaramella 2003. 24 This will include anti-cyclonic and anti-seismic building standards,
Xoraxai. Information Office in Brussels 7 National Institute of Standards and Technology 1999. which should be based on hazard and vulnerability assessment.
has compiled reports showing 8 Canada Mortgage and Housing Corporation 2005. 25 Hazard mapping is a simple and effective way of ensuring that hazards
In 1999, the United Nations that the exclusion of the Roma 9 US Census Bureau 2000. Data from the 1999 American Housing Survey. are recorded and updated on a regular basis. The maps shall cover the
Economic and Social Council from mainstream life in Europe In 2003, the American Housing Survey further found that 11 per cent of entire city and its boundaries, be available to the public and as recent
(ECOSOC) reported that the runs contrary to the norms of the total housing stock was considered as having minor technical defi- as possible (less than five years old).
majority of Roma people in European legislation and inter- ciencies.
Europe lived in the “most squalid national human rights conven- 10 In Canada, of the housing stock that needed major repairs, 39.2 per cent
and derelict housing estates” tions. In September 2002, Alvaro were houses on First Nations reserves. In the United States, the share
with sanitary facilities that were Gil-Robles, the Commissioner of households living in houses with severe problems was more than
either extremely poor on non- for Human Rights for the two times higher for blacks and Hispanic populations than white non-
existent. Since 1999, the situation has not changed much. In Council of Europe, stated that the Roma in Greece were living Hispanic populations (3.4 and 3.8 compared to 1.5). US Census Bureau
fact, according to most reports, it is getting worse. The United under conditions “very remote from what is demanded by 2000.
Nations, the World Bank, the European Commission and other respect for human dignity”. And in 2003, a World Bank report 11 In the United States, this ratio is 15.7 per cent of renter houses with
organizations have produced indicators that show that the found that “the Roma inhabit approximately 95 per cent of selected deficiencies versus 7.5 per cent of owner-occupied houses.
standard of living in Roma ghettos in Europe is appallingly the chabolas (makeshift housing and slums) around larger (US Department of Housing and Urban Development 2005.) In Canada,
low, with life expectancy rates sometimes 20 years less than cities in Spain. Approximately 80 per cent of these houses are the ratio is only 9.3 per cent versus 7.4 per cent. (Canada Mortgage and
the majority populations, unemployment rates close to 100 smaller than 50 square meters and house more than 4 people Housing Corporation 2005).
per cent, excessive school drop-out rates, and high incidence … . The lack of sanitation and running water in these areas 12 However, the Housing Statistics of Indonesia published that in 2000
of violence. Despite these indicators, such ghettos are not threatens the health of the inhabitants.” only 14 per cent of the national urban housing stock was considered
only allowed to remain, but more are being created. In fact, non-permanent. Data on floor, roof and walls requires further refine-
the construction of a wall built around one such community In Romania – which has Europe’s largest Roma population of ment. Refer to Housing Statistics Indonesia 2000.
in the Czech town of Usti nad Labem was cited by a recent approximately 2 million people (although the name Roma is 13 Planning Commission of India 2002.
report as “a single but vivid example of such ghettoization in not derived from Romania) – persecution of the Roma has 14 Observers and researchers doubt that the situation has evolved posi-
all its shocking reality”. been occurring since the mid-1800s. During World War II, the tively in recent years due to economic uncertainty and additional expen-
Nazis also slaughtered between 500,000 and 1,500,000 Roma ditures on security matters. US Library of Congress 1998.
Roma settlements in Central and Eastern Europe have during the holocaust. 15 For countries with information on the three components, regression
increased, as have the number of Roma who are forced to live equations using Demographic and Health Survey data show that the
in them. Most affected by the collapse of the Communist According to Valeriu Nicolae of the European Roma percentage of durable housing is strongly correlated with the percent-
economies, evicted because they cannot afford market rents, Information Office in Brussels, although more than 20 years age of houses with durable flooring material in Africa, Asia and Latin
the Roma often have no other choice but to settle in have passed since the European Parliament discovered in its America and the Caribbean. For the three regions the regression equa-
makeshift housing located on land no one else wants: con- 1984 report that the Roma in Europe faced societal and legal tions are statistically similar; that is, one regression equation from
taminated industrial properties or garbage sites isolated from discrimination, the 8 to 12 million Roma living in Europe are these three data sets is sufficient to estimate the durable housing from
the majority population, without public utilities such as clean still considered third-class citizens and there are few, if any, the nature of the floor whatever the region.
drinking water, electricity or waste collection. concrete actions targeting the improvement of their living 16 UN-HABITAT 2004a.
conditions. If the situation continues unabated, it is likely that 17 Martínez-Martín 2005.
In Western Europe, slum-like settlements housing Roma com- the Millennium Development Goals will not be achieved 18 UN-HABITAT 2005c.
munities can be found in Greece, Portugal and Spain. In Italy among this much-neglected community. 19 In countries such as Bangladesh, the prevalence of non-durable hous-
ing is around 50 per cent. UN-HABITAT 2005c.
Source: Nicolae 2005. 20 Economic and Social Commission for Asia and the Pacific (ESCAP) 2003.

68 69
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
2.2 Not Enough Room: Overcrowding in Urban Households
“This one room is my bedroom, my kitchen and my sitting room.”
Jared Odhiambo, a slum dweller in Kibera, Nairobi.1
H
aving only one room for sleeping, eating
and socializing may be perfectly fine for a
single person on a tight budget, but imag-
ine managing a family of four, five or even
more in an area fit for just one. Renting a
squalid, overcrowded one-roomed house or apartment is the
only way many low-income families around the world can
afford shelter at all in urban areas. Family members must
adapt small spaces to suit their daily needs, often at the cost
of their privacy and health.
Recent studies of connections between housing conditions
and rates of illness and child mortality have contributed to
the growing realization that good-quality housing conditions
are essential to ensuring a healthy, productive population.
The risk of disease transmission and multiple infections
increases substantially as the number of people crowded into
small, poorly ventilated spaces increases. A study on over-
crowding in low-income settlements conducted by UN-
HABITAT in 1995 confirmed that infectious diseases are
likely to thrive in overcrowded and low-income households,
owing to lack of ventilation, lack of hygiene and exposure to
environmental contaminants.2 The prevalence of overcrowd-
ing in inadequate dwellings has also been linked to increases
in negative social behaviours, such as domestic violence and
child abuse, and to negative outcomes of education and child
development. Children’s education may be affected by over-
crowding directly, owing to a lack of space to do homework
and the disruption of sleep patterns, and indirectly, through
absenteeism caused by illness arising in part from overcrowd-
ing.3 Research has also suggested that overcrowding may lead
to the eviction of some tenants, since congestion increases
the likelihood of property damage and may violate rental
agreements.
In cities of the developing world, overcrowding in low-
income areas is often related to other forms of social and
physical deprivation. It is not just a question of parents shar-
ing a bedroom with their grown-up sons and daughters, or
too many people sleeping in the same room; rather, as one
inhabitant living in a Nairobi slum put it, overcrowding
“takes one’s dignity away”. Living in crowded quarters inten-
sifies interpersonal contact and the experience of sights,
sounds and smells – often for the worse. At the community
70
level, high residential densities can put excessive strains on
social services, such as medical clinics, and on schools, natu-
ral environments and other resources.
Overcrowding is a manifestation of housing inequality that
results from a combination of factors. Insufficient housing
stock and lack of affordable housing are perhaps the most
prominent factors leading to overcrowding, coupled with
market and policy situations unfavorable to low-income res-
idents: inefficient housing markets, inadequate public and
private investment in affordable rental units and inappropri-
ate design of available units, among others. From a structur-
al perspective, unemployment, lack of living-wage jobs and
the spatial concentration of ethnic minorities and people liv-
ing in poverty are significant factors.
■ Global trends in overcrowding
Scholars have argued that overcrowding is a hidden form
of homelessness. People without an adequate place to live are
often forced to search for accommodation with friends or rel-
atives. This can place stress on the hosts, whether tenants or
owners, making accommodation for visitors and kin even
more tenuous. Consequently, many people occupy dwellings
that exceed local standards of occupancy. In Chile, “los alle-
gados” (people living with other families) today represent
slightly more than one-fourth of the country’s urban popula-
tion.4 Chileans are clearly coping with the national housing
deficit through co-habitation instead of creating new slums.
In Australia, 58 per cent of the homeless population takes
shelter with friends or relatives for sometimes six months or
longer, surpassing the “chronic” homelessness threshold in
that country.5 In Haiti, as in many other parts of the devel-
oping world, many individuals and families “time share” the
same house, occupying it in shifts.
In developed countries, overcrowding as a physical housing
problem has substantially decreased over time. According to
the UN-HABITAT definition (see box), which is based on
conditions in developing countries, overcrowding is non-
existent or extremely rare in most countries (less than one-
half of one per cent of the urban population). In Amsterdam,
for example, data collected by the Netherlands Department
Beyond Culture: Defining Overcrowding
The perception of overcrowding is subject to
cultural definitions and is often a function of
standard dwelling unit sizes, family groupings
and other cultural norms. In some cultures, and
among some ethnic groups, living in close quar-
ters is preferable to living in smaller family
groupings, or it is at least tolerated.
Behavioural studies indicate that certain levels
of crowding are desirable among some groups.
In the United States, for instance, 8 per cent of
high-income Asian and Hispanic groups contin-
ue to live in houses considered “overcrowded”
by American standards, even though they can
afford to enlarge their living space or move to a
larger house. Non-Hispanic and non-Asian peo-
ple with incomes comparable to those groups
experience overcrowding only half as often.
Even though it is difficult to place a value judg-
ment on overcrowding, given its cultural speci-
ficity, household surveys suggest that if given
the choice, very few people would be willing to
share a bedroom with four or five people.
Culture, in terms of tolerable crowding levels,
has some limits in that sense.
Sources: Myers & Baer 1996; www.stats.govt.nz; UN-HABITAT 2002b.
Many people, especially the poor, have few
options regarding whether or not to live in
crowded spaces. Overcrowding, therefore, can
be an important indicator of substandard hous-
ing, whereas sufficient living area is a key indi-
cator for measuring adequacy of shelter. In this
sense, overcrowding can be an objective meas-
ure that transcends culture and ethnicity. As
one of the shelter deprivation indicators, over-
crowding expresses a normative judgment
about the degree of crowding, which applies a
criterion that defines a particular density as
acceptable or unacceptable.
There is no basis in scientific literature for
choosing one standard of unacceptable over-
crowding over another. Countries define the
crowding indicator in different ways. Some
developed nations apply the concept of the
adult individual’s need for a separate bedroom,
and any value in excess of 1.0 – any bedroom
used by more than one or two adults – repre-
sents a measure of crowding. Other countries
determine the number of bedrooms a dwelling
should have to provide freedom from crowding.
Mexico City MARK EDWARDS/STILL PICTURES
A common standard is defined by the number
of people per dwelling, per room or per bed-
room, with some countries separating the num-
ber of individuals by age group, in which a gen-
der disaggregation is fundamental.
Overcrowding can also be defined in terms of
the square meters available per person, in
which values are determined according to the
number of individuals.
UN-HABITAT and its partners developed an
operational definition of overcrowding as one
of the slum-related indicators: the “proportion
of households with more than two persons per
room”. This definition was developed consider-
ing that reduced space and high concentration
of people in the dwelling is often associated
with certain health risks, so may be correlated
with slum conditions. After observing the sta-
tistical distribution of more than two persons
per room throughout the world, UN-HABITAT
revised its definition to three persons per room.
As part of the UN-HABITAT monitoring exercise,
the indicator is described in the positive as
“sufficient living area”.
71
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

for Housing indicates that less than 1 per cent of the city’s
MAP 8 PROPORTION OF HOUSEHOLDS LACKING SUFFICIENT LIVING AREA, 2003
households have more than three persons per room.
However, according to the Amsterdam housing standards cri-
teria (one person per room), 26.2 per cent of houses are over-
crowded.6 In Canada, only 0.014 per cent of households
reported having more than three persons per room.
Nonetheless, if a much higher standard is applied as per the
country’s own definition, 6.3 per cent of urban households
were below the standard in 2003.7 In many developed cities,
housing overcrowding (as per national standards) is correlat-
ed with the prevalence of ethnic minorities. In London, the
ward with the highest proportion of households with more
than one person per room is Wembley Central, a ward that
also has the highest proportion of residents of Indian origin.8
In Australia, overcrowding is approximately nine times more

Apartments in Phnom Penh ©ERIK DEGRAAF. IMAGE FROM BIGSTOCKPHOTO.COM

>40%
prevalent among the aboriginal population than it is among
>25-40%
non-indigenous people.9
>15-25%
People living in cities in developing countries, however, still
<=15%
experience high levels of overcrowding. According to UN-
HABITAT urban indicators, approximately 20 per cent of the
world’s urban population was living in inadequate dwellings, in Source: UN-HABITAT 2005, Global Urban Observatory, Urban Indicators Programme, Phase III.
terms of sufficient living area, in 2003. However, the extent
and nature of this phenomenon varies among regions. In 2003,
two-thirds of the developing world’s urban population without
■ Overcrowding data by region
sufficient living area resided in Asia, half of this group lived in
Southern Asia (156 million people). Africa ranked second with
75 million people living in overcrowded conditions. Over one- Overcrowding in Africa Overcrowding in Asia
tenth of the developing world’s urban population without suf-
ficient living area (49 million people) resided in Latin America About a quarter of sub-Saharan Africa’s urban population lives In Asia, sufficient living area is poorly reported, so determin-
and the Caribbean. in overcrowded houses. Residential overcrowding is more preva- ing the actual incidence of overcrowding for the region is diffi-
lent in some cities than in others, including Addis Ababa, cult. However, some trends are beginning to emerge. Southern
Kampala and Luanda, where more than 40 per cent of the urban Asia has the highest prevalence of overcrowding in the develop-
population lives in housing that does not have sufficient living ing world, with a third of its urban population residing in hous-
TABLE 2.2.1 SUFFICIENT LIVING AREA COVERAGE AMONG FIGURE 2.2.1 DISTRIBUTION OF URBAN POPULATION LACKING
areas. Similar deficits are found in the Nigerian cities of Lagos es that lack sufficient living area, followed by South-Eastern Asia
URBAN POPULATION BY REGION, 2003 SUFFICIENT LIVING AREA, 2003
and Ibadan. Residents of Ibadan have experienced the sharpest where over a quarter of the urban population lives in overcrowd-
Distribution of decline in sufficient living area in the world: from 95 per cent of ed housing. For Eastern Asia, not enough information was avail-
urban popula-
Population tion lacking the urban population in 1990 to 70 per cent in 2003. It is likely able for analysis, and levels and trends were estimated from those
Access to Urban lacking sufficient that this phenomenon was precipitated by the disproportionate observed in South-Eastern Asia and Western Asia. Differentials
sufficient population sufficient living area in
living area, in 2003 living area developing Northern Africa
increase in the property market value during the 1990s, as a across sub-regions should therefore be examined with caution.
2003 (%) (thousands) (thousands) world (%) result of the construction of the Ibadan-Lagos expressway that An analysis of sufficient living area in this region shows that
Sub-Saharan Africa encouraged many workers in Lagos to move to Ibadan, where most of the cities and countries are facing growing trends toward
accommodation is cheaper. This influx raised the housing overcrowding. The few exceptions include the cities of Manila in
Latin America and the Caribbean
demand and increased prices, simultaneously forcing the urban the Philippines, and the cities of Istanbul, Ankara and Adana in
Northern Africa 90.5 77,910 7,429 1.9 Eastern Asia poor to find rooms in the cheapest areas of Ibadan – the inner Turkey, where urban residents slightly increased their living area.
Sub-Saharan Africa 73.1 251,166 67,629 16.8 city and peripheral slums – and increasing overcrowding rates.11 Overcrowding rates are high in various Asian cities, namely in
Southern Asia In Northern Africa, the incidence of overcrowding is 10 per
Latin America and Yangon, in Myanmar, Dhaka and Rajshahi in Bangladesh,
the Caribbean 88.2 417,229 49,176 12.2 South-Eastern Asia
cent of the urban population. However, some countries, notably Karachi, Faisalabad and Islamabad in Pakistan, and Ulan Bator
Eastern Asia 91.5 564,871 47,813 11.9 Egypt, have dramatically reduced overcrowding in the last two in Mongolia, where around 40 per cent of the urban population
Southern Asia 65.0 448,738 156,849 39.1 Western Asia decades. The cities of Cairo, Alexandria, Port Said and Suez lived in overcrowded dwellings in 2003. With the exception of
South-Eastern Asia 73.1 228,636 61,448 15.3 increased the proportion of their inhabitants with sufficient liv- Myanmar, which has the lowest proportion of slums among the
Western Asia 91.1 124,370 11,111 2.8 ing area from 70 per cent in 1990 to 95 per cent in 2003. In the least developed countries in the world (26 per cent), the other
Total 401,456 Moroccan cities of Casablanca and Rabat, the percentage of peo- nations are characterized by a high prevalence of urban dwellers
ple with sufficient living space rose from 69 per cent and 79 per living in slum conditions: Bangladesh, at 85 per cent; Pakistan,
Source: UN-HABITAT (2006), Urban Indicator Programme Phase III and United Nations, World Urbanization Prospects; The 2003 revision. cent, respectively, in 1998, to 79 per cent and 87 per cent, at 74 per cent; and Mongolia, at 65 per cent. In other countries
Note: Access to sufficient living area was computed from Demographic and Health Surveys (DHS) data. respectively, in 2003. in which the proportion of slum dwellers is high, such as Nepal

72 73
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

(56 per cent) and India (55 per cent), the incidence of over-
crowding is also relatively high – one-third and half of the urban
population, respectively.12
Overcrowding in Latin America and the Caribbean
In Latin America and the Caribbean, information on suffi-
cient living area is lacking for most of the Caribbean countries
and many South American nations. Estimations were made
using data from countries that represent just over 50 per cent
of the entire region’s population. However, estimates indicate
that whereas the region has made significant progress in
improving slums, overcrowding affects over 10 per cent of the
urban population.
The highest levels of overcrowding in the region are found
in Central America, particularly in Guatemala and Nicaragua
(30 and 38 per cent, respectively). Both countries had a high
prevalence of slum households in their cities in 2001, at more
than 60 per cent, and among the highest rates of slum growth
countries, with more than two-thirds of their urban popula-
tions living in slum conditions
Overcrowding in the developed world
Studies carried out in developed countries confirm that over-
crowding affects some specific populations more than others.
Robust research evidence corroborates the fact that tenants
are more likely live in overcrowded units than homeowners.13
In the United States, for instance, overcrowding is approxi-
mately twice as prevalent among tenants as among owners.
Likewise, households made up of young occupants are more
likely to be overcrowded than households comprised of older
adults, and higher rates of overcrowding are found among
recent immigrants than other residents. Hispanic and Asian
communities account for 8.3 per cent of all households in the
United States, but they represent 46.6 per cent of all over-
crowded households.14
In cases in which it has been possible to adjust for con-

Kolkata, India RON GILING/STILL PICTURES

in the region between 1990 and 2001 – 2.4 per cent and 3.4
per cent per year, respectively – more than twice the average
slum growth rate of the rest of the region (1.3 per cent). In
South America, overcrowding rates were higher in Bolivia and
Peru, where around one-third of the urban populations were
deprived of sufficient living space in 2003. This is not surpris-
ing, considering that slums remain a major challenge in both
founding variables, such as tenure status, income and age, the
conclusions are clear: overcrowding is one of the most signif-
icant ways in which urban poverty expresses itself. Statistical
analysis conducted by UN-HABITAT shows that overcrowd-
ing does not appear as an isolated variable, but it is very often
combined with other slum dimensions, such as lack of safe
drinking water and improved sanitation.

FIGURE 2.2.2 PROPORTION OF URBAN HOUSEHOLDS WITH FIGURE 2.2.3 PROPORTION OF URBAN POPULATION WITH
SUFFICIENT LIVING AREA ACCESS TO SUFFICIENT LIVING AREA IN SELECTED CITIES, 2003

100 100

80 80

60 60
Endnotes
Percentage (%)

Percentage (%)

1 Quoted in Phombeah 2005. 10 In sub-Saharan Africa, information was obtained from countries repre-
40 40
2 UN-HABITAT 1995. senting more than 80 per cent of the region. In Latin America and the
3 United Kingdom 2005. Caribbean, data was derived from countries representing slightly more
4 Ministerio de Vivienda y Urbanismo 2004. “Allegados” in Chile may be than 50 per cent of the region. Lack of information was observed on this
20 20
a hidden form of slum-dwellers. If that is the case, national data should indicator in some sub-regions of Asia; for example, in Eastern Asia none
be revised. of the countries reported information on sufficient living area, and in other
5 Australian Bureau of Statistics 2001. sub-regions, this information was obtained from only a few countries.
0 0
6 Netherlands Department for Housing 2005. 11 Fouchard 2003.
n

ba

sa

'a

on

i
ric n

ric n

be nd

ia

ia

As rn

ia

ba

ch
Fe
ja

ol

ak
Af her

Af ara

na
As

As

As
e

ha
ba

ng
rib a a

7 Canada Mortgage and Housing Corporation 2005. 12 Other countries that are characterized by having a high prevalence of
id

Pa

um

ra
a

an

ia
st

Dh
Sa
rt

Ab

ns
sA

Ya

Ka
n

rn

rn
Ea
Ca ric
Sa

M
o
No

er

Ki
he

te

Sa
h-

di
th Ame

st
b-

es
ut

slums are Afghanistan (98 per cent) and Cambodia (72 per cent); no infor-
ut

Ad

8 London Research Centre 2005.

Ea
Su

W
So

So
tin
e
La

9 Australian Bureau of Statistics 2004. In some regions, such as mation currently exists on sufficient living area for these countries.
1990 2003 Africa Latin America Asia Warburton, overcrowding stood at 50 per cent for indigenous house- 13 See, for instance, Myers & Baer 1996; and Ellaway & Macintyre 1998.
and the Caribbean holds and 2 per cent for non-indigenous households. 14 Myers & Baer 1996.
Source: UN-HABITAT Urban Indicators Programme Phase III.
Note: Data for 1990 not available for some regions. Source: UN-HABITAT 2005, Urban Indicators Programme Phase III.

74 75
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

2.3 Safe Drinking Water in Cities

U
nited Nations statistics on safe drinking water
provision throughout the world indicate a
slight improvement in recent years: between
1990 and 2002, approximately 1.1 billion
people gained access to an improved source of
drinking water, an increase in global coverage from 77 per
cent to 83 per cent.1 Access to safe drinking water is, howev-
er, unevenly distributed around the globe. Significant dispar-
ities exist on several levels. First, safe drinking water is
unevenly distributed between the urban and rural popula-
Indonesia PETRUS IYAY SAPUTRA/UNEP/STILL PICTURES

tions of the world: 95 per cent of the world’s urban dwellers

have access, but only 72 per cent of the world’s rural popula-

Cebu City MARK EDWARDS/STILL PICTURES

tion has access.2 Second, per capita water consumption levels
vary widely between rich and poor nations, with the former
consuming 10 times more drinking water than the latter: 500
to 800 litres per day, compared with 60 to 150 litres per day.3
Third, asymmetries exist in water access and water manage-
ment between high-income countries experiencing low popu-
lation growth and low-income countries facing rapid popula-
tion growth and water scarcity problems. And fourth, there
are extreme differences in the quantity and quality of water
that rich and poor households can obtain within the same city
in different parts of the world. some countries, people suffering from waterborne diseases Drinking water supply is often intermittent even in neigh-
Some of the disparities in drinking water provision are occupy a high proportion of hospital beds – as in India, where FIGURE 2.3.1 WATER EXPENSE AS A FRACTION OF HOUSEHOLD bourhoods where people have water pipes and taps inside their
clear, but others – particularly intra-city differences in access 65 per cent of hospital patients are being treated for water- INCOME IN ADDIS ABABA, 2003 houses. In the coastal city of Mombasa in Kenya, a study found
and consumption – are less evident, as they are often dis- related illnesses at any given time.7 Many people also spend 20
that very few parts of the city had continuous water supply; on
guised by aggregated urban data that averages out quantity large proportions of their incomes on the treatment of water- average, water was available for less than 3 hours per day, and
and quality of water among those having access to safe water related diseases. In sub-Saharan Africa, for instance, people in some parts of the city, pipes had not been functional for sev-
and those who are frequently deprived. This results in a sin- living in poverty spend at least one-third of their incomes on eral years.10 In other instances, such as in Dhaka, Bangladesh,
gle, and misleading, estimate of access to safe drinking water treatment of water-borne and water-related diseases, such as 15
slum dwellers complain that the agency that provides water

Ratio of water expense to income (%)

in a city.
Current United Nations statistics, which use aggregated
data, confidently report that 95 per cent of the world’s urban
population has “improved” water provision (see box). At least
12 countries with low or middle income levels even report
that they have 100 per cent coverage.4 More than 15 coun-
tries that perform poorly on a number of health indicators
linked to living conditions also report that their national
water coverage is above the world’s urban average (95 per
cent)5; and 44 countries (18 in Africa, 12 in Asia and 14 in
Latin America and the Caribbean), each with slum popula-
tions representing at least one-fourth of their total urban pop-
ulations, report that water coverage is almost universal – as
high as 90 per cent.6
Yet, millions of people living in these countries suffer from
waterborne diseases, indicating that they do not have ade-
quate access to safe drinking water as officially reported. In
diarrhoea and malaria.8 The facts suggest that the drinking
water available to people living in poverty is frequently inad-
equate or contaminated.
Although most statistics reflect better drinking water cover-
age in urban areas than in rural areas, various surveys show
that in many cities, the quantity and quality of water available
to low-income residents falls short of acceptable standards.
Hundreds of millions of people who supposedly have access
to water only have access to communal pipes with intermit-
tent water supply shared by hundreds, or even thousands, of
people. In the lowest-income areas, people also pay more for
water than their wealthier neighbours. UN-HABITAT urban
indicators show that in 2003 in Jakarta, for example, only 29
per cent of the households were connected to piped water in
their dwellings; 19 per cent used private wells as a second
source of water; and 7 per cent purchased water from vendors,
who charged several times more than the official price.9
supply and sewerage only makes functional water connections
available to land owners, not tenants in slums.11
Water is, therefore, not as safe as it appears according to sta-
10 tistics, nor is it supplied in high enough quantities to all
households, thus compelling many families to find more
water from other sources that are less reliable. Differing defi-
nitions, technical measurement problems and political
5 unwillingness to report accurate data prevent the existing
water access issues among people in cities from being cap-
tured for statistical purposes. The need to acknowledge differ-
ential access and the problems associated with it is urgent:
0 nearly half of the urban populations in Africa, Asia, and Latin
First Second Third Fourth Fifth America suffer from one or more of the communicable dis-
(Poorest) (Richest) eases associated with inadequate water and sanitation provi-
Source: UN-HABITAT, Addis Ababa Urban Inequity Survey 2003.
sion.12 Too often, the myriad issues associated with lack of
Note: Based on wealth of households, households divided equally into five groups, safe drinking water go unrecognized by the development
First (Poorest), Second, Third, Fourth and Fifth (Richest). community. Clearly, however, lack of safe water kills. The

76 77
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

“Improved” or “Adequate” Access to Water: Definitions and Issues of Measurement

As part of its strategy to monitor slum improvement,
UN-HABITAT included inadequate access to safe
drinking water as one of five shelter deprivation
indicators. Lack of access to improved water facili-
ties and lack of access to improved sanitation are
coincident with the other indicators of slum house-
holds – housing durability, overcrowding and secure
tenure – and together account for the identification
of most slum households.
To measure access to safe drinking water, UN-HABI-
TAT adopted the definition of “improved” water sup-
ply developed by the WHO/UNICEF Joint Monitoring
Programme for Water and Sanitation. The definition
is widely accepted and is backed by a long history
of data collection in developing countries.
Definition: The water should be affordable and at
sufficient quantity that is available without exces-
sive effort and time.
Indicator: The proportion of households with
access to improved water supply with:
• Household connection
• Public standpipe shared by a maximum of two
households
• Borehole well
• Protected spring
• Rainwater collection
At least 20 liters per person should be available
within an acceptable timespan (defined locally). This
is a variation from the original WHO/UNICEF defini-
tion, which indicates that water should be available
within one kilometer of the residence; the definition
was changed to accommodate data collection in
high-density urban areas. Some data collectors
count all residents with a water source within 200
meters of their home as having access to water, but,
as UN-HABITAT has found in further research, that
having a tap within 200 meters of a dwelling in a
rural settlement with 200 people using it is not the
same as having a public tap within 200 meters of a
dwelling in an urban squatter settlement with 5,000
people using it.
The WHO/UNICEF definition was designed to meas-
ure water access in rural contexts and does not nec-
essarily provide a suitable definition for research in
urban areas. Urban settlements have particular
needs for water that are distinct from rural areas,
yet it is still common to refer to “improved” and
“adequate” access to water interchangeably in both
urban and rural settlements. These terms cannot
capture the full extent of two different realities:
rural and urban.
“Improved” water provision is often no more than a
public tap shared by several hundred people with an
intermittent supply of water. Definitions of services
and access vary not only among the different types MAP 9 PROPORTION OF HOUSEHOLDS WITHOUT ACCESS TO IMPROVED WATER, 2003
of surveys undertaken, but also over time. It is
therefore sometimes difficult to compare surveys
undertaken even within the same country. In addi-
tion, people often use more than one water source,
and it is difficult to ascertain the quality, accessibili-
ty, reliability, and cost of each, and whether its use
is a problem.
United Nations estimates do not as yet consider the
quality and affordability of water services. UN-
HABITAT is in the process of refining survey ques-
tions on access to improved water through the
application of Urban Inequities Surveys, within the
framework of the Monitoring Urban Inequities
Programme. The new questions will appear on
household surveys regularly administered by UN-
HABITAT’s partner agencies.
Gender issues in safe water access, child access to
facilities and the number of households using the
same facility are other important issues to consider <5%
>5-12%
in urban water research. UN-HABITAT is now work-
>12-50%
ing with the WHO/UNICEF Joint Monitoring
Programme to harmonize and standardize indicators >50%
and methodologies, in order to match definitions
and methods of measurement to produce data com- Source: UN-HABITAT 2005, Global Urban Observatory, Urban Indicators Programme, Phase III.
parable across countries and cities.

root of this unrelenting catastrophe is well known13: people cent); in Africa, 7 per cent rely on manual pumps, and in
are not getting sufficient quantity and quality of water that is Latin America and the Caribbean, 5 per cent use them.
affordable and available without having to invest excessive TABLE 2.3.1 IMPROVED DRINKING WATER COVERAGE AMONG FIGURE 2.3.2 DISTRIBUTION OF URBAN POPULATION LACKING
effort and time. URBAN POPULATION BY REGION, 2003 IMPROVED DRINKING WATER BY REGION, 2003
UN-HABITAT data on urban indicators collected in 2003 ■ Global trends
provides the distribution of households by major source of
drinking water at the national, urban agglomeration and The world has made important progress in increasing access to Distribution of
slum/non-slum levels. Analysis of the data at each level reveals safe drinking water. Despite this increase, however, the total num- urban popula-
Urban Population tion lacking
that getting water from a tap is a luxury enjoyed by only two- ber of people who gained access to improved water sources Access to safe Population lacking safe safe water in
third of the world’s urban population. In 2003, 62 per cent of remained stable at approximately 17 per cent, owing to global water source, 2003 water developing Northern Africa
2003 (%) (thousands) (thousands) world (%)
all city dwellers had access to piped water, 46 per cent of population growth.15
Sub-Saharan Africa
whom had water piped into the dwelling and 16 per cent of Data collected by the WHO/UNICEF Joint Monitoring
whom had a water tap in the yard or plot. Public taps serviced Programme provides additional information on access of the Northern Africa 94.9 77,910 3,960 2.4 Latin America and the Caribbean
10.4 per cent of the urban residents, and 8 per cent had access world’s urban population to safe drinking water. The share of the Sub-Saharan Africa 82.0 251,166 45,210 27.6
to manually pumped water or protected wells.14 urban population with access to improved water sources remained Latin America and Eastern Asia

UN-HABITAT data reveals important inter-regional differ- stable from 1990 to 2002 at 95 per cent. The Joint Monitoring the Caribbean 95.2 417,229 20,166 12.3 Southern Asia
ences in the way urban residents in the developing world gain Programme counts access to all types of improved water sources Eastern Asia 92.5 564,871 42,365 25.9
access to safe water. In 2003, Latin America and the that are protected from external contamination, including piped Southern Asia 94.3 448,738 25,428 15.5 South-Eastern Asia
Caribbean has the highest proportion of urban households household connections, public standpipes, boreholes, hand-dug South-Eastern Asia 91.0 228,636 20,577 12.6
Western Asia
dependent on piped water sources (89.3 per cent) and sub- wells, springs, and rainwater collection.16 However, the number Western Asia 95.1 124,370 6,115 3.7
Saharan Africa has the lowest (38.3 per cent). In Africa, 20 of people without access to improved water will double between Total 163,822
per cent of the urban population cites public water taps as a 1990 and 2010, increasing from 108 million to 215 million.
primary source of drinking water – twice the world’s average Widespread inter-regional differences in water access exist,
– while in Latin America and the Caribbean, only 2 per cent though these asymmetries began to conform around higher cov-
of residents depend on public taps. Asia has the highest pro- erage in the 1990s, particularly in urban areas. An outline of water Source: UN-HABITAT 2005, Urban Indicator Programme Phase III and United Nations, World Urbanization Prospects; The 2003 revision.
portion of people using manually pumped water (12 per access trends in the major regions of the world follows. Note: Access to safe water was computed from data of WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation.

78 79
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

Phnom Penh MIKE KOLLOFFEL/STILL PICTURES

■ Water in Africa
Africa has the lowest safe drinking water coverage, with only
64 per cent of the total population having access to an improved
water supply – that, is water that is affordable and of sufficient
quantity, and available without the investment of excessive time
or effort. The situation is much worse in rural areas, where cov-
erage is only 50 per cent, than in urban areas, where coverage is
about 85 per cent.17 In global terms, 29 per cent of the world’s
urban population without access to improved water supply lives
in Africa.
In 2003, 82 per cent of sub-Saharan Africa’s urban population
had access to an improved drinking water supply – a smaller pro-
portion than urban dwellers in any developing region in the
world. In 2002, only 45 per cent of rural sub-Saharan Africans
had access to safe water. The sub-region as a whole did not expe-
rience any changes in the coverage levels during the 12 years
between 1990 and 2002, remaining at 82 per cent throughout.18
In some countries, however, urban populations are grossly under-
served. In Cape Verde, Eritrea, Niger and Rwanda, only one-
third of the urban population had access to an improved water
source in 2001.
Lack of access to piped water affects some cities within the sub-
region more than others. Some capital cities have had particular-
ly low coverage, including Kampala, Uganda, at 15.1 per cent,
and Kigali, Rwanda, at 35 per cent. Moreover, the proportion of
households having access to piped water in Central and Western
African capital cities is much lower than the sub-region’s average,
particularly in Luanda, Angola (13.1 per cent); Yaoundé,
Cameroon (33.5 per cent); Ouagadougou, Burkina Faso (33.8
per cent); and Conakry, Guinea (39.2 per cent). Access to piped
water in some secondary African cities is even worse: in Nazret,
Ethiopia, the proportion is 16 per cent; and in Butembo,
FIGURE 2.3.3 ACCESS TO PIPED WATER IN SELECTED CITIES IN
SUB-SAHARAN AFRICA, 2003
40
35
30
Access to piped water (%)
Democratic Republic of the Congo, the proportion is 14.4 per
cent.19 Lack of water source improvement is bound to continue
affecting health, education, productivity, and income generation
throughout the sub-region.
In Northern Africa, access to safe water in urban areas was
almost universal in 2002, at 96 per cent. In this sub-region, the
urban population is expected to increase slowly and the number
of people living in slums should decline. It is more likely that the
reduction in the number of slum dwellers will aid the sub-
region’s performance on the other slum indicators, particularly
overcrowding and sanitation, than on the water indicator, as
other shelter deprivations are more prevalent than access to safe
water. Additional government efforts are required to reach the
remaining 4 per cent of the population that lacks access to an
improved water source. Those who still lack safe drinking water
may be more difficult to reach because their communities are
physically isolated from others and their pervasive poverty limits
their ability to improve their own facilities.
Data on piped water connections in urban areas reveal con-
trasting histories in Africa as a whole. In Northern Africa, piped
water coverage increased from 83 per cent in 1990 to 96 per cent
in 2003, due in large part to significant improvements in
Egyptian and Moroccan cities. Despite an acute shortage of
water in the sub-region, most residents of its cities and rural areas
enjoy regular, affordable piped water access. The data from
Northern Africa supports the theory that the world’s “water cri- es 79 per cent in urban areas. Impressive growth has taken
sis” is more of a political and governance crisis than a physical place in several Indian cities, namely in Hyderabad, Amritsar, TABLE 2.3.2 CITIES MAKING RAPID PROGRESS IN DRINKING
scarcity crisis.20 Akola and Hisar, where access to improved water sources WATER COVERAGE, 1999 -2003
increased from 48 per cent to 86 per cent between 1990 and
2003. The most impressive gains were recorded in the city of Cities that increased coverage by
at least 15% between 1990 and 2003
■ Water in Asia Hai Phong, Viet Nam, which provided improved water
sources to twice as many people from 46.4 per cent of the
Asia accounts for two-thirds of the world’s population that urban population in 1990 to 99.6 percent in 2002. Drinking Water
Coverage (%) % increase
lacks safe water: 670 million people in both rural and urban City/Country 1990 2003 1990-2003
areas.21 In urban areas, coverage is over 90 per cent in most
sub-regions, but poor access to water facilities in urban areas ■ Water in Latin America and the Caribbean
Kigali, Rwanda 72.7 84.5 16
is reported in various countries, particularly in Cambodia,
where only 58 per cent of people living in cities have access to Water coverage estimates based on data from the Ouagadougou, Burkina Faso 77.5 94.3 22
an improved water source and Lao People’s Democratic WHO/UNICEF Joint Monitoring Programme suggest that 95
Republic, where 66 per cent have access. The most impressive per cent of the urban population and 69 per cent of the rural Fortaleza, Brazil 68.9 93.7 36
gains in access to improved water sources were made in population in Latin America and the Caribbean had access to
Goiânia, Brazil 85.7 98.8 15

Southern Asia’s cities fuelled primarily by increased coverage improved drinking water sources in 2002. Today, around 20
25
in India from 88 per cent in 1990 to 96 per cent in 2002. million people in the region’s urban areas are without access to Tijuana,Mexico 67.8 98.2 45
20 The proportion of the population with access to an improved water supply. While most countries in the region have
improved water source in urban areas in Asia is very high – 93 over 90 per cent coverage in urban areas, some countries have Agartala, India 79.2 99.6 26
15
per cent – as reported by the water utilities and ministries in relatively low coverage: Anguilla (60 per cent), Argentina (85
Bitung, Indonesia 73.6 97.7 33
10 charge of drinking water services. Piped water is more pre- per cent), Belize (83 per cent), Dominican Republic (83 per
cious, since only 70 per cent of the Asian urban population cent), Ecuador (81 per cent), El Salvador (85 per cent), Jamaica Cebu, Philippines 66.5 98.3 48
5
had access to it in 2002, approximately half of whom had (81 per cent), Haiti (49 per cent), Panama (88 per cent), Peru Da Nang, Viet Nam 84.7 96.6 15
0
access to a working tap within the dwelling. The heterogene- (88 per cent) and Venezuela (88 per cent). In Haiti less than half Jaya Pura, Indonesia 47.1 94.1 100
Co . o

ne y

da i

ol a

O iop th

Fa ou

oo é
nd la

an al
ui kr
e ep b

ng nd

er nd
th re

of the urban population has access to an improved water source.

ga pa

w Kig

ity of the region is made clear by the contrasts observed in the

th m.R tem

a ug
ng of

a)

a)

(B uag ia)

n)
(G ona

a)

so
(A ua

am u
(E aza
(U am

kin do
o)
e u

Metro Manilla, Philippines 83.8 96.9 16

C

(C oaY
(D B

ur a
K

level of piped water coverage in the different sub-regions. By Figures on piped water in the region are available only for
(R

far the lowest proportion of people having access to piped major cities in a few countries: Brazil, Colombia and Rajahmundry, India 83.6 99.6 19
Source: UN-HABITAT, Global Urban Observatory, Urban Indicators Programme Phase III. connections in urban areas is found in South-Eastern Asia, Guatemala. On average, urban areas in the three countries had
Note: Data based on Demographic and Health Surveys in various years. with just 45 per cent, whereas coverage in Western Asia reach- some of the highest levels of piped water connections in the Source: UN-HABITAT 2005, Urban Indicator Programme Phase III.

80 81
 MILOON KOTHARI
WATER AND SANITATION: ONLY A HUMAN RIGHTS APPROACH WILL DO
Economic globalization policies – part of a global structural adjustment
agenda that finds its most boisterous proponents among the wealthy
nations of the world – have lent momentum to an ongoing movement
toward privatization and commodification of basic services, such as water
and sanitation. This phenomenon, now widely assumed to be irreversible,
coupled with the inability of governments to provide their citizens with
affordable access to such services, tends to have a disproportionately
severe impact on those most vulnerable segments of the population, the
poor and socially marginalized.
Water, essential to human life and all life on the planet, is part of the glob-
al commons and arguably the most quintessential of all collective
resources. It is not a private commodity to be bought, sold or traded for
profit – an exclusive luxury accessible to a few and elusive to the majority.
This fundamental principle is clearly articulated in the General Comment
No.15 (2002) of the UN Committee on Economic, Social and Cultural Rights,
which says that:
“Water should be treated as a social and cultural good,” and that “invest-
ments should not disproportionately favour expensive water supply servic-
es and facilities that are often accessible only to a small, privileged frac-
tion of the population, rather than investing in services and facilities that
benefit a far larger part of the population.”
The human rights of people and communities to housing, water and sanita-
tion – long recognized as indivisible, and guaranteed under international
law – continue to be eroded as the processes of privatization become more
entrenched and quicken in pace. While the promise of economic globaliza-
tion to help alleviate want and reduce poverty may exist in the abstract, its
basis on the Washington Consensus and reliance on a theory of presumed
trickle-down benefits find little basis in history. The time has come to
rethink current global economic and social policies, and the perverse and
brutalized neo-liberal logic that underpins them, and reaffirm our commit-
ment to the human rights principles and standards that offer the only real
paradigm for improving the lives of millions of the poor.
The consequences of having inadequate or no access to water, while uni-
versally devastating, tend to be more acutely felt by women and children.
When water is not readily available, it is principally women and children
who are charged with the burdensome responsibility of its collection, often
assessing the impact of globalization on not only the right to adequate hous-
ing in particular, but in a broader sense on the extent of States’ compliance
with their legal obligations under various international and human rights
treaties and guidelines. By transforming a basic social service and scarce
resource into an economic commodity, the world’s economic and policy plan-
ners are operating under the myopic macroeconomic assumption that exist-
ing water resources can be managed and consumed efficiently in accordance
with competitive market principles. Let us not be naïve. A consideration of
the three major criticisms of privatization will readily dispel any notion that
the basis of such an assertion is to be found in reality: private businesses put
too much emphasis on profits and cost recovery; services to vulnerable
groups are inadequate and of poor quality; and private operators are not
accountable to the public. What’s more, the lack of capacity, or willingness,
on the part of States to regulate the operations of private providers only mag-
nifies the above outlined shortcomings of privatization.
There is an acute need to strengthen participatory monitoring mechanisms,
as processes of privatization are extremely difficult to reverse once imple-
mented, and corporations enjoy formidable legal recourse through multilat-
eral trade agreements. Consequently, the expansion of any such agree-
ments, such as the World Trade Organization General Agreement on Trade
Services (GATS), which led to the privatization of social services and the
entry of corporations into the arena of providing social goods such as water,
will only serve to exacerbate an already adverse situation. The right to an
effective remedy for anyone whose rights have been violated cannot be con-
tracted away by the State nor denied by the operations of intergovernmen-
tal institutions. Investment or trade bodies should not adjudicate concerns
that fall firmly within the ambit of human rights as if they were simply dis-
putes between corporations and state actors. Any violation should, and
must, be dealt with through the relevant human rights enforcement mecha-
nisms that seek the integration of human rights obligations into national and
international policy making, thereby establishing a clear and positive prece-
dent for the future.
Of equal or greater importance is the need for sustained vigilance at each
stage of this protracted campaign, to actively safeguard against the collater-
al erosion of other human rights during the ongoing effort to achieve the
Goals. The effort at improving the living conditions of some, by way of slum
upgrading projects, for example, must not lead to the breach of human rights
of others, such as through forced eviction or the now rampant phenomenon
developing world in 2003, with the Colombian cities of
Bogotá, Medellin, Neiva, and Valledupar reporting universal Endnotes
coverage. Slightly less than two-thirds – 71 per cent – of the
households included in the data had piped connections in their 1 WHO/UNICEF Joint Monitoring Programme for Water Supply and
homes, with the remaining third using a tap in the yard or plot. Sanitation 2004.
2 Ibid.
3 UNEP 2003.
■ Understanding access to safe drinking water in 4 These countries are: Andorra, Barbados, Belize, Botswana, Chile,
urban areas: The case of Addis Ababa Costa Rica, Dominica, Egypt, Lebanon, Paraguay, Ukraine, and
Zimbabwe. (WHO/UNICEF Joint Monitoring Programme for Water
UN-HABITAT is working on refining the methodology used Supply and Sanitation 2004.)
to measure access to “improved water supply”. An urban 5 These countries are Albania, Armenia, Azerbaijan, Bolivia, Colombia,
inequities survey conducted in Addis Ababa, Ethiopia, showed Côte d’Ivoire, Dominican Republic, Gabon, Guatemala, Honduras,
that if the indicator includes variables for measuring the propor- Kazakhstan, Malawi, Namibia, Pakistan, Sri Lanka, and Uzbekistan.
tion of people with access to safe water that is affordable, provid- (WHO/UNICEF Joint Monitoring Programme for Water Supply and
ed in sufficient quantity and does not require excessive time or Sanitation 2004).
effort to acquire, the number of urban dwellers without sufficient 6 UN-HABITAT 2005c.
access is much higher than officially acknowledged (see figure). 7 The Virtual Water Forum Final Report, 3rd World Water Forum, March
Statistics collected by the Government of Ethiopia in Addis 2003: http://www.worldwaterforum.org.
Ababa, for instance, report that 88.5 per cent of the urban pop- 8 Hansen & Bathia 2004.
ulation has improved water provision;24 however, UN- 9 UN-HABITAT 2005c.
HABITAT’s study, as part of its Monitoring Urban Inequities 10 Rakodi, et al. 2000.
Programme (MUIP), demonstrated that the proportion of 11 Civil Society submission to the Government of Bangladesh for the
urban residents with an improved water supply can drop to 21.3 development of the “National Strategy for Economic Growth, Poverty
per cent if the operational definition that includes ease of access Reduction and Social Development”, Poverty Reduction Strategy
is combined with variables on sufficient quantity, affordability Paper. Bangladesh, 2003.
and time required to collect it.25 The same study demonstrated 12 World Health Organization 1999.
that when data is disaggregated at intra-city levels, massive dis- 13 Bartram, et al. 2005.
parities are apparent. Official statistics belie the actual condi- 14 The remaining percentage, 18.8 per cent, could not be described with
tions under which people in poverty live. In Addis Ababa, the a single variable, owing to differences in definitions and methods of
survey showed that the proportion of non-slum urban house- computation. UN-HABITAT Urban Indicators Programme Phase III,
holds with access to safe water was almost two times higher than 2005.
the proportion of slum households with access to safe water. 15 UNFPA 2003.
16 UN-HABITAT 2005c.
FIGURE 2.3.4 ACCESS TO WATER DECREASES DRAMATICALLY WHEN 17 WHO/UNICEF Joint Monitoring Programme for Water Supply and
COST AND QUALITY ARE CONSIDERED: THE CASE OF ADDIS ABABA Sanitation 2000. The percentage reported by on water supply cover-
age in urban areas is 85 per cent, while according to UN-HABITAT’s
100
urban indicators, this percentage is 89 per cent.
18 UN-HABITAT 2005c.
80
19 Secretariat of the Third World Water Forum, 2003.
20 Hansen & Bathia 2004.
Percentage (%)
60

expending inordinate amounts of time and energy in the process. This has of land-grabbing in all of its forms. 21 WHO/UNICEF Joint Monitoring Programme for Water Supply and
a detrimental impact on their health, security and education. While the lack 40 Sanitation 2000.
of sanitation facilities affects both men and women alike, sanitation needs A human rights approach must both inform the normative discussion, as well 22 UN-HABITAT 2005c.
20 23 WHO/UNICEF Joint Monitoring Programme for Water Supply and
and demands tend to differ as a function of gender. Women have particular as guide the processes surrounding efforts to achieve the Millennium
needs and concerns of privacy, dignity and personal safety, and the lack of Development Goals. Such a strategy, coupled with existing international Sanitation 2004.
0
sanitation facilities in the home can force women and girls to use secluded human rights treaties, declarations and guidelines, provides a framework 24 Addis Ababa City Council 2004.
Ke da
a
Ak B ka
Ne i Ka e
fa liti
ilk

et a
a
Ak B ka
e
fa liti
ilk

et a
a
Ak B ka
Ne i Ka e
fa liti
ilk
m

ak ol

s K ad
em

ak ol

s K ad
em

ak ol
sS

sS

sS
Ye

Ye

Ne Ka

Ye
is Ara
te

di Ar 25 UN-HABITAT 2004b.

di Ar
places, often at great distance from the home, thereby exposing them to through which the formulation of responsible economic policies for the ben-

i
d
Ad

Ad

Ad
heightened risk of sexual abuse. Furthermore, lack of accessible basic serv- efit of humankind can become a reality.
ices can often lead to or further exacerbate tense and stressful relations By neighborhood in Addis Ababa, Ethiopia
within the home, increasing women’s vulnerability to domestic violence.
Access to improved water
Access to improved water when cost is factored in
Privatization of water and sanitation services warrants close scrutiny when Miloon Kothari is the United Nations Special Rapporteur on Adequate Housing. Access to improved water when cost and quality are factored in

Source: UN-HABITAT, Addis Ababa Urban Inequity Survey 2003.

83
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

2.4 The Silent Tsunami: Improved Sanitation: A Basic Principle

The High Price of Inadequate Sanitation in Urban Areas A household is considered to have access to
“improved” sanitation if it has a human
excreta disposal system, either in the form of
a private toilet or a public toilet shared by a
maximum of two households. In urban areas,
access to improved sanitation is defined by
direct connection to a public, piped sewer;
direct connection to a septic system; or access
to pour-flush latrines or ventilated improved
pit latrines, allowing for acceptable local
technologies.
Sanitation is not simply about disposal of
human waste. It is a broad concept that
encompasses the safe removal, disposal and
management of solid household waste,
wastewater, industrial waste and the like.
“Improved” sanitation, however, refers to a
basic sanitation approach that focuses on
securing sustainable access to safe, hygienic
and convenient facilities for human excreta
disposal. The first, and least expensive, step
toward a more broadly integrated sanitation
system in any community, improved sanitation
includes the need for privacy, safety, hygiene
and convenience.

■ Linking inadequate sanitation and health
Globally, an estimated 2.6 billion people lack toilets and
other forms of improved sanitation. UN-HABITAT analyses
reveal that while the world’s cities have made significant
progress in improving people’s access to water, access to
improved sanitation lags far behind, particularly in sub-
Saharan Africa, Southern Asia and Eastern Asia, where the pro-
portion of the urban population having access in 2003 was only
55 per cent, 67 per cent and 69 per cent respectively.
Lack of access to an adequate toilet not only violates the dig-
nity of the urban poor, but also affects their health. Urban
poverty is often related to poor hygiene – the result of inade-
quate sanitation facilities combined with an inadequate or
unsafe water supply. Every year, hundreds of thousands of peo-
ple die as a result of living conditions made unhealthy by lack
of clean water and sanitation options. The number of deaths
attributable to poor sanitation and hygiene alone may be as
high as 1.6 million per year – five times as many people who
died in the 2004 Indian Ocean tsunami. Inadequate sanitation
is therefore something of a “silent tsunami” causing waves of ill-
Ecuador T. NEBBIA/UNEP/STILL PICTURES
eases, particularly among the sick and children.4 In 77 per cent
of the 144 cases in the WHO study, positive health benefits in
the form of reduced incidence of diarrhoea could be definitive-
ly attributed to improvements in sanitation systems, whereas
only 48 per cent could be attributed to improvements in the
water supply. Another multi-country study confirmed that it is
possible to obtain a reduction of up to 37 per cent in cases of
diarrhoea when access to improved sanitation facilities is pro-
vided to unserved populations.5
Lack of sanitation is not always a top priority among people
living in poverty, whose needs for drinking water and sufficient
food often take precedence. Poor sanitation is perceived as
“tomorrow’s priority” even if it has life-threatening conse-
quences. Thus, slum dwellers in some African cities resort to
disposal of excreta in plastic bags, which then get discarded
carelessly in drainage channels, rubbish bins or in the streets of
the neighbourhood itself. These so-called “flying” toilets in
Nairobi or “mobile” toilets in Lusaka have negative health con-
sequences that are not often apparent to the inhabitants. Low
priority accorded to sanitation is also manifested by the fact
that a limited number of households having latrines make

ness and death, especially among children. Although
Millennium Development Goal 7, target 10, aims to halve the
proportion of people without access to safe drinking water and
basic sanitation by 2020, the relationship between water and
sanitation is very often ignored when allocating resources; for
every dollar invested in water supply, only 20 cents goes toward
the provision of basic sanitation.1 This explains the huge gap
between water and sanitation coverage in the world: 83 per
cent of the world’s residents have access to safe drinking water,
appropriate use of them, and an even smaller number maintain
them properly. For instance, a study conducted in Zambia,
Zimbabwe and South Africa showed that only 17 per cent of
the population maintain their latrines properly.6 There is clear-
ly a strong need to link sanitation and hygiene to education,
awareness-raising and cultural attitudes toward waste disposal.
Attributing unhygienic practices only to lack of concern
about inadequate sanitation, however, can misguide analysis
and conclusions. Poverty and deprivation also play key roles.

S
ince 350 B.C., when the Greek philosopher Aristotle
posited a distinction between the public sphere of
political activity and the private sphere associated
with family and domestic life, debates about the
right to privacy have dominated popular discourse.
In today’s world, privacy is increasingly determined by individ-
uals’ power and social status: the rich can withdraw from soci-
ety whenever they wish, but those living in poverty cannot so
easily escape their neighbours’ gaze. This is particularly true
with regard to sanitation. People living in poverty are subject to
intrusion and observation around their most private affairs and
habits; they are much less likely than their wealthier neighbours
to have access to safe sanitation facilities behind closed doors.
The absence of decent toilets in impoverished neighbourhoods
violates residents’ right to privacy and is an affront to their dig-
nity. Being deprived of adequate sanitation facilities is the most
direct and most dehumanizing – but least often acknowledged
– consequence of poverty.
More than 25 per cent of the developing world’s urban pop-
ulation lacks adequate sanitation.
but only 58 per cent have access to improved sanitation.2
Research has made it clear that those without access to ade-
quate sanitation are more exposed to diseases than other groups
experiencing lack of safe water and other shelter deprivations.
They are 1.6 times more likely to experience diarrhoea, and
they have consistently higher rates of morbidity and mortality.3
This has been amply demonstrated by several studies, includ-
ing a 1996 study by the World Health Organization (WHO)
in Asia, which concluded that sanitation and hygiene are
among the most influential factors in reducing diarrhoeal dis-
For instance, residents of the neighbourhood of Mbare in
Harare, Zimbabwe, prefer to defecate wherever possible in the
open because their pour-flush toilets are overused and poorly
maintained. In that community, up to 1,300 people share one
communal toilet with only six squatting holes, most of which
are no longer flushable.7 In Nepal, two-thirds (67 per cent) of
the country’s population defecates in the open despite the fact
that half of them have access to latrines; they consider the exist-
ing latrines unsanitary and unsafe.8 In many countries, children
are afraid or reluctant to use latrines because they are perceived

84 85
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

Defining Sanitation: To estimate or to underestimate, that is the question.

Getting a clear picture of the global sanitation sit-
uation is complicated by the variety of terminolo-
gies used for data collection. An individual or a
household can have access to “adequate” or
“improved” sanitation depending on the definition
used. Studies indicate that between 850 and 1.13
billion people worldwide lacked “adequate” sani-
tation, whereas about 400 million people lacked
“improved” sanitation in 2004. These figures refer
to the sanitation hardware, or technologies, used,
including sewerage systems, toilets and hygienic
latrines. More accurate estimations should also
consider the software, or the hygiene conditions,
as well, such as maintenance of latrines and provi-
sion of hand-washing facilities.
Monitoring adequate or improved sanitation is fur-
ces and can be used by all family members, includ-
ing children over the age of 5 years.
Differences in definitions and data reporting meth-
ods make direct comparisons difficult. Some coun-
tries have higher standards for defining adequate
sanitation services, excluding ordinary pit latrines,
or counting only ventilated improved pit latrines or
flush toilets connected to a septic tank or sewer-
age system. In Uganda, for example, pit latrines are
counted as sanitary, and the latest Demographic
and Health Survey shows 80 per cent of house-
holds with access. But, if pit latrines are not count-
ed, the population with access shrinks to a mere 3
per cent. These variations in standards make the
quality of the data questionable in some countries.
For instance, in Zambia, one of the least developed
that fall into disrepair. This explains why some of
the data from individual countries shows rapid and
implausible changes in coverage from one assess-
ment to the next.
UN-HABITAT is working with the WHO/UNICEF
Joint Monitoring Programme to standardize defini-
tions to allow for more accurate global compar-
isons. Based on UN-HABITAT suggestions that
most of the simple pit latrines and traditional
latrines in use in cities are in fact unsanitary, the
data has been revised for 2004. Figures for people
without adequate sanitation have thus changed
from 2.4 to 2.6 million worldwide. The
WHO/UNICEF Joint Monitoring Programme agreed
to reduce by half the share of the population with

Manila HARTMUT SCHWARZBACH/UNEP/STILL PICTURES

ther complicated by the fact that “access” assumes
the use of the facilities; yet, in many locations,
facilities are not being used. A family may have
access as defined for purposes of the indicator, but
may fail to use the facility for practical, cultural, or
social reasons. Defining “use” of the facilities,
therefore, can be nuanced to “convenient use”. In
Nepal, for instance, the government considers a
“functioning latrine” as one that is kept free of fae-
countries, more than 87 per cent of the population
has access to adequate sanitation, while Brazil, a
middle-income country that is far wealthier and
much more developed than Zambia, reports access
of only 83 per cent. Furthermore, data from surveys
and routine reports can also have significant dis-
crepancies, because some government reports rely
on outdated data that fails to take into account
new informal settlements and sanitation facilities
adequate access who use traditional, pit or simple
latrines. For instance, the Government of Burkina
Faso reported that 54 per cent of the urban popu-
lation had access to improved sanitation. However,
out of this percentage 84 per cent, had access to
pit latrines, only half of which (42 per cent) were
counted as improved sanitation. These changes
were made in order to facilitate inter-regional
comparisons across time.

Sources: Hansen & Bathia 2004; UK Parliamentary Office of Science and Technology; Bendohmme & Swindde 1999; UNICEF 1997; UN-HABITAT 2005c.

as dark, dirty, unsafe, or smelly. In many places, latrines are not
available at all. Today, one out of every three children in the
developing world do not have access to a toilet of any kind in
the vicinity of their dwellings.9
Sanitation, hygiene and health are interconnected, linked by
three main factors in low-income communities: lack of access
to a safe, decent toilet; lack of awareness about the connections
between defecating in the open and contamination of food and
water sources; and cultural indifference to using the public
environment for the disposal of human waste. The combined
issues of extreme deprivation, lack of education and poor qual-
ity and maintenance of sanitation facilities are compounded by
the fact that interventions, if any, are typically neither sustained
nor systematic, owing to inadequate economic and financial
sector policies, poor urban management, and lack of political
will to respond to the needs of people living in poverty. Non-
recognition of informal settlements by the authorities in charge
of provision is also an issue in some cities.
■ Global trends in sanitation provision
In 2002, nearly half of the population of the developing
world – about 2.5 billion people – did not have access to ade-
quate sanitation. Of those lacking adequate sanitation, 76 per
cent (1.98 billion) lived in Asia; 18 per cent (470 million) lived
in Africa; and 5 per cent (130 million) lived in Latin America
and the Caribbean (see box on definition).10
As with most other shelter provisions, sanitation coverage is
significantly higher in urban areas than in rural areas. Data col-
lected by the WHO/UNICEF Joint Monitoring Programme for
Water Supply and Sanitation indicates that in 2002, 81 per cent
of the world’s urban residents had access to improved sanitation,
compared to 37 per cent in rural areas. As expected, these values
are lower for the developing regions, where the proportion of
the population with adequate sanitation in urban areas is 73 per
cent, and in rural areas is 31 per cent.11 Estimations of the
deficit depend heavily on the data sources and definitions used.
While 73 per cent of the urban population in the develop-
ing world has access to adequate sanitation facilities, more than
560 million city dwellers are still deprived of a basic, decent
toilet facility. The region with the lowest coverage of improved
sanitation in urban areas is Africa (63 per cent), with a sub-
regional variance of 55 per cent coverage in sub-Saharan Africa
at one end, and 89 per cent in Northern Africa at the other end.
At the continental level, Africa accounts for around one-fifth
(21 per cent) of the world’s population lacking improved sani-
tation. Asia has the second-lowest coverage in urban areas in the
world (66 per cent). Because of the population sizes of China and
India, along with other large nations in the region, Asia also
accounts for the largest numbers of people without sanitation
nearly 400 million people. Afghanistan has by far the lowest pro-
portion of the urban population with access to improved sanita-
tion (16 per cent) in the region.12
Latin America and the Caribbean has a relatively high sanita-
tion coverage of 84 per cent; the region is performing better on
sanitation than Asia and Africa, and it hosts only 12 per cent of
the world’s population without access to improved sanitation.
More than 75 per cent of the urban populations in virtually all
countries in the region have improved sanitation, with the excep-
tion of Haiti and Belize – two countries in which less than 50 per
cent of the urban population has improved sanitation facilities.
In contrast, people in the developed world enjoy more or
less universal provision of advanced sanitation facilities.
Virtually all households have access to improved sanitation, as
more than 98 per cent of homes are connected to piped,
municipal sewage treatment systems. Those not connected to
municipal sewage systems use septic tanks and similar solu-
tions that are regulated, inspected during construction and
regularly tested for performance. In Japan, for instance, 100
per cent of the population has improved sanitation: 81 per
cent of the population is connected to municipal sewerage sys-
tems, and 19 per cent have household septic tanks (johkasoh)
or use other improved facilities in agricultural communities.13
North America has the highest reported coverage for any
Sanitation, hygiene and
health are interconnected
world region as a whole: 100 per cent. Reporting mechanisms
for Europe are extremely poor – contrary to what might be
assumed – so available data for that region is not statistically rep-
resentative.14 Available data indicates that 99 per cent of the
European population has access to improved sanitation, but in
many of the region’s new economies, the infrastructure for san-
itation still needs to be developed or improved. Consequently,
more than 80 million people in the region, or 10 per cent of the
total population, do not have improved sanitation.15
■ Urban sanitation data for regions of the develop-
ing world
Based on the definition of improved sanitation, and taking into
consideration the limitations of measurement, it is possible to
claim a modest improvement in the proportion of the world’s
urban residents with access to sanitation: from 68 per cent in
1990 to 73 per cent in 2003.16 A significant proportion of the
urban population in the developing world consequently does not

86 87
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

use or have access to any type of sanitation facility; an estimated
10 per cent of the world’s urban population, or 300 million peo-
ple, defecate in the open or use unsanitary bucket latrines.17
The problem with these estimates is that different countries
use different types and categories of latrines, making it difficult
to code them for monitoring purposes. Apparently, not all of the
reported facilities are improved; where they are, a large propor-
tion of them are overcrowded, unsafe, lack provisions for clean-
ing hands and body, and are poorly maintained. In many urban
settings, especially in densely populated areas, pit latrines do not
significantly reduce the risk of faecal-oral diseases because of
their unhealthy conditions. Studies have shown that latrines
alone do not have a clear health impact unless the behavioural
patterns associated with sound hygiene practices are also
ensured. In Pakistan, for instance, the fact that only a negligible
difference exists in the frequency of diarrhoeal episodes between
households having latrines and those without latrines indicates
that hygiene is inadequate.18
Sanitation in African cities
In Africa, the increase in the proportion of people with
improved sanitation varies greatly, depending on the sub-region.
A positive trend has been observed in Northern Africa, whereas
signs of stagnation are clear in sub-Saharan Africa. In the former
sub-region, the population enjoying improved sanitation services
grew from 84 per cent in 1990 to 89 per cent in 2002, due pri-
marily to an increase registered in the Egyptian cities of Cairo,
Alexandria, Port Said and Beni Suef, and the Moroccan cities of
Casablanca and Rabat. Yet, small cities did not experience the
same growth in provision; some still have access comparable to
TABLE 2.4.1 IMPROVED SANITATION COVERAGE AMONG URBAN
POPULATION BY REGION, 2003
Distribution of
urban popula-
Population tion lacking
cities in sub-Saharan Africa, where performance on sanitation
MAP 10 PROPORTION OF HOUSEHOLDS WITHOUT ACCESS TO IMPROVED SANITATION, 2003
indicators is as poor as on other development indicators.
Centralized sewerage systems are the most appropriate solution,
in terms of expense, infrastructure and health returns. In
Northern Africa, increased sewerage provision is the most com-
mon solution in large, primary cities – national capitals and eco-
nomic centres. In the sub-region’s secondary cities, however,
piped sewerage technology is not as common; in Assyut, Egypt,
fewer than 20 per cent of the households are connected to piped,
waterborne sewer systems.
In 2003, slightly more than half of the urban population in sub-
Saharan Africa enjoyed adequate sanitation facilities (55 per cent)
– roughly the same proportion that was recorded in 1990. The rate
of urbanization in this sub-region has been the highest in the
world: it experienced an 80 per cent increase in the number of
urban residents between 1990 and 2003. Subsequently, the num- >50%
ber of people lacking improved sanitation has drastically increased >25-75%
>10-25%
– from 77 million in 1990, to 132 million in 2001, to 160 million
<=10%
in 2005.19 If no remedial action is taken, poor sanitation will con-
tinue to have a significant – and dismal – impact on the lives of
Source: UN-HABITAT 2005, Global Urban Observatory, Urban Indicators Programme, Phase III.
people living in poverty in the cities of sub-Saharan Africa.
Evidence of success exists in a few cities that have managed to
expand coverage between 1990 and 2003 significantly, including Sanitation in Asian cities Southern Asia’s coverage is also among the highest, with an
Kigali, Rwanda, and Ibadan, Nigeria which increased coverage increase of 12 percentage points, although it started with the
from 47.8 per cent and 26.8 per cent in 1990 to 79.4 per cent and Time series analysis based on UN-HABITAT urban indica- lowest baseline in Asia – 54 per cent in 1990. Several Indian
67.3 per cent in 2003, respectively. Some countries, such as tors shows that Asia has made major progress on the provision cities – namely Akola, Kanpur and Kharagpur – made rapid
South Africa and Zimbabwe also have extensive sewerage net- of improved sanitation in cities. The region is diverse and het- progress, increasing coverage by at least 25 per cent between
works. This example gives hope and direction to the achievement erogeneous, however, with both advanced and poor economies. 1990 and 2002. However, Southern Asia also has countries with
of sanitation target of the Millennium Development Goals. In South-Eastern Asia experienced the highest growth in low proportions of urban dwellers with access to improved san-
contrast, sanitation facilities in some cities such as Addis Ababa in improved sanitation coverage since 1990, increasing from 67 itation: Nepal (68 per cent); and Afghanistan (16 per cent).20
Ethiopia and Porto Novo in Benin have less than half of their per cent that year to 79 per cent in 2002, particularly in mid- Growth in access to improved sanitation in Eastern Asia
population served with improved sanitation. sized cities such as Bogor and Kediri in Indonesia, Cagayan de was rather moderate, increasing from 64 per cent in 1990 to
Oro in Philippines, and Hai Phong in Viet Nam. About 29 69 per cent in 2003, owing in large part to the increase in
million people still lacked access to improved sanitation in China and Mongolia’s largest cities.21 Deficits in sanitation
FIGURE 2.4.1 DISTRIBUTION OF URBAN POPULATION LACKING 2005; by 2020, the deficit is expected to decrease to 25 million. facilities in the two countries remain high: in China, 33 per
IMPROVED SANITATION BY REGION, 2003
cent of the urban population still lacks improved sanitation,
as does 54 per cent of the urban population of Mongolia.
In Western Asia, coverage was quasi-universal in 1990, at
96 per cent, but since then, it has been difficult for countries

Guanabara Bay, Rio de Janeiro ©ELDER SALLES. IMAGE FROM BIGSTOCKPHOTO.COM

Access to Urban lacking improved
improved Population improved sanitation in Northern Africa to reach the poorest of the poor. In some of the middle-
sanitation, 2003 2003 sanitation developing
(%) (thousands) (thousands) world (%) Sub-Saharan Africa
income countries of the sub-region, a great deal of the infra-
structure is in place, but much of it is in poor condition and
Latin America and the Caribbean does not function reliably.22
Northern Africa 89.4 77,910 8,245 1.5
Eastern Asia
The sanitary situation of the countries of the
Sub-Saharan Africa 55.1 251,166 112,815 20.1
Commonwealth of Independent States in Asia is not well
Latin America and
Southern Asia documented. According to data from the WHO/UNICEF
the Caribbean 84.2 417,229 66,061 11.8
Eastern Asia 69.4 564,871 172,756 30.8 South-Eastern Asia
Joint Monitoring Programme, there are significant disparities
Southern Asia 67.0 448,738 148,084 26.4 in access to improved sanitation between urban and rural
South-Eastern Asia 80.0 228,636 45,727 8.2
Western Asia areas: 81 per cent and 45 per cent, respectively. Multiple
Western Asia 94.9 124,370 6,322 1.1 Indicator Cluster Surveys conducted by UNICEF in the
Total 560,011 region reveal that the only available government statistics are
limited to the proportion of the population served by cen-
Source: UN-HABITAT (2006), Urban Indicator Programme Phase III and United Nations, World Urbanization Prospects; The 2003 revision. tralized sewerage systems.23 Since such systems are generally
Note: Access to safe water was computed from data of WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation. limited to central districts of the region’s large cities, smaller

88 89
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

cities are not normally covered, and only 10 per cent of Despite improvements in the overall quality of life in some been recorded in cities such as Guatemala City, from 33 to 85
FIGURE 2.4.3 THE UNEQUAL DISTRIBUTION OF IMPROVED
households in rural areas are covered.24 Almost all the popu- developing countries, many cities, such as Rondonopolis in per cent between 1990 and 2003. The city of Fortaleza in Brazil
SANITATION IN ROSARIO, ARGENTINA
lation not served by centralized sewerage systems uses individual Brazil and Guayaquil in Ecuador, have been largely unsuccess- offers a valuable lesson in how development of sanitation infra-
sanitation facilities, such as pit latrines and pumped latrines. ful in creating reliable and adequate sanitation services. Studies structure can have a positive health outcome. The city experi-
Observations made in various parts of the region suggest that suggest that in most smaller urban centres, the proportion enced significant reduction in infant mortality rates from 74
most pit latrines, for individual household use and in public without adequate sanitation provision is even higher; most per 1,000 births to 28 per 1,000 births in 2001 - the same peri-
institutions such as schools and hospitals, are in poor condi- urban centers in low- and middle-income nations have no sew- od in which sanitation coverage increased from one-third to
tion, especially in economically disadvantaged areas. ers at all and have little or no other public support for good- more than half the urban population.
quality sanitation.26 However, significant improvements have

Sanitation in Latin America and the Caribbean

Latin America and the Caribbean has relatively high sani-

tation coverage but with vast internal variations between Endnotes
rural and urban environments, and among individual coun-
1 From 1990 to 2000, sanitation received only 20 per cent of the $16 bil- 20 UN-HABITAT 2005c.
tries and cities. According to data compiled by the
lion invested in water by national governments and external support 21 WHO/UNICEF 2004.
WHO/UNICEF Joint Monitoring Programme, the propor-
agencies. (United Kingdom Parliamentary Office of Science and 22 Ibid.
tion of the rural population lacking improved sanitation was
Technology 2002.) 23 According to UNICEF, 54 per cent of the population in five of the
almost half of the urban proportion: 44 per cent versus 84
2 WHO/UNICEF 2004. Commonwealth of Independent States countries has access to central-
per cent, respectively. In various countries of the Caribbean, 0% - 22% 3 Evans 2005. ized sewerage systems.
improved sanitation coverage is universal, particularly in 23% - 50% 4 South-East Asia Regional Office of WHO, quoted in Evans 2005. 24 Cherp 1999.
Antigua, Bahamas, Barbados, Cuba, Surinam, and Trinidad 51% - 71% 5 Esray 1996. 25 WHO/UNICEF 2004.
and Tobago.25 UN-HABITAT urban indicators data reveals 72% - 90% 6 Manase, et al. 2001. 26 Satterthwaite, et al. 2005.
high coverage in various cities, namely in Bogotá and 91% - 100%
7 Ibid.
Medellin in Colombia; Caracas in Venezuela; Chillan in 0 0.9 1.8 Kilometers 8 Conference Proceedings for the South Asian Conference on Sanitation,
Chile; and Guadalajara in Mexico. Source: Javier Alberto Martinez-Martin, 2005. Dhaka, Bangladesh, 2004.
9 UNICEF 2004.
10 WHO/UNICEF 2000 assessment.
FIGURE 2.4.2 PROPORTION OF URBAN HOUSEHOLDS WITH FIGURE 2.4.3 CITIES WITH LOW SANITATION COVERAGE: 11 WHO/UNICEF 2004.
ACCESS TO IMPROVED SANITATION PROPORTION OF URBAN POPULATION WITH ACCESS TO 12 WHO/UNICEF 2000 assessment.
IMPROVED SANITATION, 2003 13 Report on Urban Indicators submitted by the Japanese Government to
100
UN-HABITAT, May 2005. Refer as well to the Report on Urban Indicators
60
prepared by the Canada Mortgage and Housing Corporation 2005.
14 In the WHO/UNICEF 2000 assessment, the coverage data represented
50 15 per cent for 1990 and 44 per cent of the region’s total population. For
80
the UN-HABITAT urban indicators, no reporting mechanism was devel-
40
oped, with the exception of that by the Dutch Government.
60 15 WHO 2005.
Percentage (%)

16 Data is based on Urban Indicators Programme, Phase III. The percent-

30 age in 1990 is the average of 29 cities in Africa, 54 cities in Asia and
40 25 cities in Latin America and the Caribbean. The percentage in 2003 is
20
the average of 55 cities in Africa, 78 cities in Asia and 31cities in Latin
America and the Caribbean.
20 17 As mentioned earlier, a large proportion of people with latrines prefer
10
to defecate in the open. A study conducted by the World Bank in
Mumbai concludes that toilets for slum dwellers, even when available
0 0 are in such poor condition that people refuse to use them. (Nitti &
Sarkar 2003).
ric n

ric n

be nd

ia

ia

As rn

ia

ba

da

Og seru

ag sho

em s
ab
ui
kr

i
Af her

Af ara

ak

go

ut

go

ol
As

As

As
e

aq
rib a a

ba

an

Re
a

an

ia

na

ap
st

óp
m

o
ou

ou
a
rt

ay
n

rn

rn

sA

Lu
Ea

18 Conference Proceedings, South Asian Conference in Sanitation,

Ba
Ca ric

Co

M
Sa

on
No

er

Dj

ad
bo
he

te

Gu
h-
th Ame

di
st
b-

nd

Si
es
ut

ut

Ad
Ea
Su

Ro
W

Ou
So

So

Pakistan, 2004.
tin
e
La

19 Estimations based on the UN-HABITAT Urban Indicators Data 2005 and

1990 2003 Africa Latin America and the Caribbean
the United Nations Urbanization Prospects 2004.
and Asia
Source: UN-HABITAT Urban Indicators Programme Phase III.
Note: Data for 1990 not available for some regions. Source: UN-HABITAT 2005, Urban Indicators Programme Phase III.

90 91
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Sanitation: A women’s issue
No issue touches the lives of
women – particularly poor urban
women – as intimately as that of
access to sanitation. In low-
income settlements where there
are no individual toilets, women
have to queue for long periods to
gain access to public toilets;
some have to bear the indignity
of having to defecate in the
open, which exposes them to the
possibility of sexual harassment
or assault.
Although men also suffer from
the burden of poor sanitation,
they are more likely to resort to
other means to relieve them-
selves. In many slums, men uri-
nate and defecate along railway
tracks and in open spaces. But
women – whose anatomy, modesty and susceptibility to
attack does not allow them to discreetly relieve themselves in
public – have no choice but to wait until dark, usually early in
the morning when there is less risk of being accosted. “Going
to the toilet” for these women often means squatting in a pri-
vate spot or waking up before dawn to queue at public toilets.
One woman interviewed in a Mumbai slum explained what it
means to have no toilet: “We use the toilet outside our settle-
ment, five minutes away. We have to stand in a queue for half
an hour. That is why the men all go under the bridge and only
the women use the toilets. Children also go out in the open.”
A disproportionate share of the labour and health burden of
inadequate sanitation falls on women. For women living in
slums, a long wait at the public toilet can mean that children
are left unattended, or that a household chore is delayed.
Unhygienic public toilets and latrines threaten the health of
women, who are prone to reproductive tract infections caused
by poor sanitation. For women who are menstruating, the
need for adequate sanitation becomes even more acute.
Moreover, because it is generally women who are responsible
for the disposal of human waste when provision of sanitation
is inadequate, they are more susceptible to diseases associat-
ed with contact with human excreta.
Despite all this, the sanitation crisis affecting women has not
been given a high priority on the agendas of human rights
and women’s organizations. United Nations and other inter-
Sources: Warah 2005a; UN-HABITAT 2003a; Hardoy, et al. 2001; Mukherjee 2001.
92
LENA SOMMESTAD
STRONGER ACTION IS NEEDED TO ACHIEVE THE SANITATION TARGET
A sanitation target was set at the World Summit on Sustainable
Communal water tap in Soweto, Johannesburg RASNA WARAH
Development, held in Johannesburg in 2002. Sanitation was also discussed
during the 2005 UN-HABITAT Governing Council and during the 30th session
of the Commission on Sustainable Development in New York in April 2005. I
am pleased that sanitation has finally attracted the attention of policymakers
worldwide. However, to achieve the sanitation target, we need stronger
action and more innovative approaches. In particular, women’s role and inter-
ests must be fully taken into account.
The present growth of the world population is almost entirely in the urban
centres of the developing world. Solutions for improved urban sanitation
therefore require urgent innovation in urban areas. However, sanitation infra-
structure planning rarely takes into account a gender perspective. Women
and children are hardly ever considered priority target groups for sanitation
provision. This needs to be improved. The social context of sanitation and the
needs of families must be built into solutions if they are to be sustainable.
Women, along with children, are the ones who suffer most from lack of water
and sanitation. Girls face lack of sanitation in schools. This is a significant
barrier to education that must be identified and removed. Women normally
have the main responsibility for work tasks associated with health and sani-
national bodies tend to confine women’s issues to reproduc- tation, such as water collection, washing and cleaning of the house and the
tive health and education. Few, if any, governments focus on latrines. Women, children and elderly people are disproportionately affected
the impact of inadequate sanitation on women. This could by illness caused by polluted water.
also be partially explained by the fact that improving access
to sanitation was only recently recognized as a pressing The challenge surrounding equity and access to sanitation is closely linked to
internationally agreed target – in 2002 at the World Summit the fact that sanitation systems are seldom planned using the elements of
on Sustainable Development – so the issue has not been on sustainability and ecological principles. This is why sustainable sanitation
the public agenda for long. Although women’s lack of access approaches have been developed.
to water in both rural and urban areas and its health impli-
cations – including severe back pain caused by carrying The essential features of sustainable sanitation are proper containment
heavy vessels of water over long distances – has been the allowing for sanitation and recycling, closing of the nutrient and water loops,
subject of several studies, women’s lack of access to sanita- protection of downstream health and environment, local management and
Kibera, Nairobi THIERRY GEENEN, FOR NAIROBI RIVER BASIN PROJECT, UNEP.
tion has not received the same attention. Preliminary UN- financing, affordability, and equitable services for rich and poor, women and
HABITAT analyses indicate the need for further study of the men, old and young. Sustainable sanitation is pursued in very close coopera-
issue, as they show that lack of sanitation in slums increases tion with the people who will use these systems. Advanced solutions are
health risks among all slum residents, women and children coupled with local knowledge.
in particular.
The Government of Sweden contributes to the development of an ecological
Because rural women – no matter how poor – do not have to sanitation approach. We are working through SIDA – the Swedish
face the same dilemma as their urban counterparts when it International Development Cooperation Agency – to support the develop-
comes to sanitation, poverty reduction efforts, which are ment of ecological sanitation in several developing countries. The Swedish
currently focused on rural areas, particularly in sub-Saharan support aims to create a global confidence in ecological sanitation as a reli-
Africa and Asia, do not factor in women’s access to sanitation able, cost-effective and sustainable alternative to conventional systems.
in urban areas. Most rural households have access to at least
one toilet – even if it is a crude pit latrine – which means that For decades, sanitation has been neglected in many parts of the world. This
women in rural areas rarely queue to go to the toilet and are has had severe consequences, in particular for the poor, and in particular for
less likely to share toilets with dozens of other people. They Today millions of poor people suffer from diseases and parasites women. Now, we have a great challenge ahead. Sanitation should be a
are also more likely to keep the toilets clean, as their family’s because sanitation has not received enough political attention. human right. Sustainable solutions are key. By empowering women and chil-
health often depends on it. Young children die of diarrhoea and women are denied security, dren, we can make a difference.
privacy and dignity. Approximately 2.6 billion people in the world
are lacking sanitation services. Lena Sommestad is the Swedish Minister for Sustainable Development.
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
2.5 Owners without Titles:
Security of Tenure in Cities of the Developing World
■ Evictions: The most severe consequence of
insecure tenure
Mass evictions of slum dwellers in various parts of the devel-
oping world in recent years have raised fears that security of
tenure and housing rights are becoming increasingly precarious
in the world’s cities. A global survey1 in 60 countries found that
6.7 million people had been forcibly evicted from their homes
between 2000 and 2002, compared with 4.2 million people
between 1998 and 2000.2 Some experts have described the
unprecedented rise in the number of evictions in the last five
years as a global “epidemic”.3
Although forced evictions are an extreme consequence of inse-
cure tenure, their increasing prevalence in recent years point to
trends that suggest that attitudes of local and national govern-
ments towards the urban poor are becoming increasingly intoler-
ant. This can be attributed to to a variety of factors, including
globalization, which is putting pressure on national and local
governments to “beautify” or “clean up” their cities in order to
become more competitive in a global economy that has seen the
gap between the rich and poor widen and dramatically increased
the price of urban land, pushing lower-income groups to the
edge of cities to unplanned and poorly serviced areas.
Evictions are particularly prevalent in sub-Saharan Africa and
Asia; most are carried out to make room for large-scale develop-
ment projects and infrastructure, such as dams or roads, or to
accommodate city “beautification” programmes. Cities that have
experienced mass forced evictions in recent years include Beijing,
Lagos, Abuja and Nairobi. Even when evictions are “justified” –
as when they are carried out in the public interest, to build roads
or other infrastructure necessary for urban development or when
they are carried out in order to “protect” slum dwellers from haz-
ards – they not do not take place in conformity with the rules of
international law.4 Most evictions are carried out without legal
notice and without following due process. Evicted people not
only lose their homes (in which they have invested a considerable
portion of their savings), they are often forced to relinquish their
personal belongings as well.
It is not uncommon for evicted families to sleep out in the
open around the demolished site without food or basic ameni-
ties. Children and women are particularly vulnerable in such sit-
uations. Incidents of rape and killing of victims during and after
eviction exercises have been reported in many places.
When evictions take place, they not only destroy homes, but
also entire communities, which can lead to urban unrest and
insecurity. Evictions result in loss of income and disrupt highly
94
integrated and complex networks of the informal economy.5
Media and other reports suggest that the magnitude of urban
evictions is currently highest in sub-Saharan African cities,
although rural evictions are also quite common. This could be
partly because urban populations in African cities have so far not
been able to organize themselves politically in large enough num-
bers to be able to resist evictions or demand rights from their
governments. In addition, many African governments inherited
outdated, elitist laws from colonial powers that discriminate
against the urban poor; these policies have led to the creation of
“apartheid-type” cities, with the neighbourhoods of the rich and
the poor clearly demarcated.6
Strong civil society action in Asia and better legislation has had
a significant impact on improving the tenure status of slum
dwellers, but with pressures to “globalize” mounting, particular-
ly in the region’s more economically successful industrializing
cities, this trend could be reversed in the near future. Moreover,
escalating land and house prices in Asian cities could lead to eco-
nomic evictions as lower-income groups are pushed out of the
city simply because they can no longer afford to live there.
In Latin America, progressive slum upgrading and regulariza-
tion programmes have increased tenure security among the
urban poor, but evidence suggests that tenure security is not
reaching the most vulnerable groups. For instance, a study in
Brazil shows that poor blacks and mulattos are least likely to live
in adequate housing with secure tenure, and are most likely to
live in slums (see figure).
However, improving the tenure of existing urban populations
is not enough; measures must also be undertaken to prevent the
growth of new slums and informal settlements where tenure
security is at risk. This requires a parallel approach to increase the
supply of planned, legal and affordable land on a scale equal to
present and future demand.
■ Tenure security: The thin line between legality
and illegality
Non-emperical evidence suggests that between 30 per cent
and 50 per cent of urban residents in the developing world lack
any kind of legal document to show they have tenure security.7
Development agencies, academics and practitioners in urban
issues concur that informal growth has become the most signif-
icant mode of housing production in cities of the developing
world. In fact, gaining access to housing through legal channels
is the exception rather than the rule for most urban poor house-
Family being evicted during slum clearance, Minas Gerias, Belo Horizonte, Brazil DYLAN GARCIA/STILL PICTURES
holds.8 In many cases the majority of inhabitants live with “owners without titles” conceals the real number of people living
tenure systems that are “informal”, which means that their in informal settlements and significantly distorts figures and esti-
occupation of land and/or housing is either illegal, quasi-legal, mates reflecting the magnitude of urban dwellers who live with-
tolerated or legitimized by customary or traditional laws, which out secure tenure in cities.
can either be recognized or simply ignored by the authorities. Security of tenure is critical to the livelihood of slum dwellers
Slums – the generic term used to classify informal, illegal or and should encompass a minimum package of rights, which
unplanned settlements – are the invisible “zones of silence” on could progressively evolve towards a higher order of rights. This
tenure security. Little is known about the formal or informal formalization process can be accomplished through an incremen-
tenure systems slum dwellers enjoy – or don’t enjoy – as official
censuses and households surveys do not at present measure
tenure security as a development indicator, even though infor- FIGURE 2.5.1 PERCENTAGE OF URBAN HOUSEHOLDS WITH
mality – or “illegality” is perhaps the most significant factor in ADEQUATE HOUSING IN BRAZIL, 1992-2003
the physical and demographic growth of cities in the developing
world and is the main mechanism through which poor people in 80
cities gain access to land and housing.9
The status of slum dwellers in developing countries is made
70
more ambiguous by the fact that they are often not included in
national censuses and household surveys, which means that their
tenure is neither recorded nor guaranteed. Often, new informal 60
settlements are not enumerated, and even when their inhabitants
Percentage (%)
are included in censuses, they normally appear as “owners” of the
50
dwelling they occupy, even though surveys and studies have
shown that large proportions of slum dwellers are actually ten-
ants or are “owners without titles”. In 2003, the Inter-American 40
Development Bank estimated that around 60 per cent of the
urban poor in Latin America were home owners even though
30
very few had land or housing titles.10 In 2005, the Central
1992 1993 1995 1996 1997 1998 1999 2001 2002 2003
Statistical Bureau in Indonesia reported that “out of those that
own their home, only 32 per cent can show legitimate proof in White Brazil Total Blacks and Mulattos
the form of a certificate from the national land agency”.11 In
Nicaragua, one-third of the urban population was recorded as Source: Morais, Maria da Piedade, rapporteur Target 11, MDGs, IPEA based on
being “owners with no deed” in 2001.12 The high prevalence of PNAD microdata, IBGE, 1992 to 2003.
95
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

Monitoring Secure Tenure invest in enterprise.16 Land and housing are perceived as owners (both public and private) and tenants play a more
market assets that have the potential of generating bottom- important role in securing tenure than titling. In most cases,
UN-HABITAT defines secure tenure as the Component Indicator 1: Proof of Documentation. up economic and social rewards, thereby reducing poverty. land in both rural and urban areas is neither registered, not is
right of all individuals and groups to effective This component indicator assumes that documentation may be considered as proof of occupancy Diametrically opposed to this way of thinking is the percep- there an official title for it. UN-HABITAT estimates indicate
protection from the State against forced evic- and therefore could provide certain levels of security. In most developing countries, tenure security
tion that access to land is a fundamental human right neces- that less than one-third of land in developing countries is
tions. Under international law, “forced evic- in informal settlements is achieved incrementally over time through the accretion of various docu-
sary for a secure livelihood. accounted for in official land records and registries23 and ques-
tion” is defined as “the permanent or tempo- ments, such as utility bills, voter registration forms, ration cards and municipal tax receipts.
rary removal against their will of individuals,
Most governments and development agencies consider ten- tions regarding ownership are tackled through customary, com-
families and/or communities from the homes Component Indicator 2: Perception of security/insecurity of tenure. ants living in informal settlements as having “insecure tenure”, munal or religious laws governing land. According to World
and/or land which they occupy, without the This indicator measures the individual’s or household’s perception of their own tenure situation. It whereas “owners” are automatically classified as “secure”. Bank estimates, in Africa formal tenure extends to only
provision of, and access to appropriate forms is based on the experience and perceptions of those who are most directly confronted with the Countries from various regions, such as Bosnia and between 2 per cent and 10 per cent of all residential land.24
of legal or other protection”. reality of evictions in a country or city. Herzegovina, Cambodia, Cameroon, Côte d’Ivoire and Gabon Unfortunately, in some parts of the world, particularly in sub-
have used indicators such as “the percentage of land parcels Saharan Africa, customary law actually works against the inter-
For global monitoring purposes, UN-HABITAT The two component indicators are complemented by a diagnostic of the policy environment that having titles”, the “percentage of the population with access to ests of women, who are prevented from inheriting land or
proposes to adopt two more component indi- helps to determine the factual status of tenure security in a specific context. This is done through a property”, or the “number of households owning their lodging” property. While an increasing number of sub-Saharan African
cators to measure secure tenure at the indi- qualitative measurement of the legal, institutional, administrative and policy environment govern- to measure tenure security.17 The Economic Commission for countries have recognized women’s equal rights to land and
vidual/household level: ing security of tenure.
Latin America and the Caribbean (ECLAC) also associates property, thus complying with international human rights stan-
insecure tenure among the urban poor with lack of owner- dards and obligations, there are still some countries, such as
ship.18 This mode of coding seems to respond to ideologies and Kenya, Lesotho, Zambia and Zimbabwe, where discrimination
tal process of tenure upgrading that evolves from de facto tenure security, multiplying the opportunities to obtain capital. As the belief systems that view rights to land and property as being in customary and personal law matters (such as inheritance) is
(taking into account a variety of socially accepted norms in land land systems and markets become more stable, more complex dependant on market forces and affordability, rather than as still permitted in these countries’ constitutions.25
and housing tenure) to de jure tenure. This approach also allows products appear, such as land and credit being placed in second- fundamental rights guaranteed by governments.
governments to build technical and administrative procedures ary markets and stocks. Yet many examples from around the world show that whilst
over time and within their own resource capacity. titling has benefited many slum communities, and deserves a ■ The Challenge of Measuring Tenure in
The precarious status of land and housing tenure among slum place in tenure policies, it has not necessarily increased access Informal Urban Settlements
dwellers can also be progressively strengthened through an institu- ■ Ownership is not always the solution to credit or prevented growth of new informal settlements.
tional and social construct in which the accretion of various docu- Empirical evidence does not support the view that full titling Since many local authorities are reluctant to recognize the
ments plays a key role in the process. The process can start initial- Although ownership is typically regarded as the most secure lifts the poor out of poverty; in many cases, an incremental existence of informal growth in their cities, or are not predis-
ly with the occupant providing simple proof of occupancy, such as form of tenure, evidence from around the world suggests that approach – based on the right to a secure livelihood – has posed to address it in a systematic manner, they do not devel-
utility bills, voter registration forms, ration cards and municipal tax ownership is not the norm in both the developed and develop- proved to be more effective in the long term. In some cases, op appropriate means to measure and monitor the level of
receipts. Gradually, documentation could evolve towards more ing world, and is not the only means through which tenure large scale titling programmes can actually contribute to legit- informality in urban areas. In fact, in 2005, UN-HABITAT’s
consolidated forms of occupancy rights, and eventually to formal security can be achieved. Home owners are a minority in most imizing and exacerbating unequal systems of land and proper- Urban Indicators Programme found that around 100 cities in
tenure regularization and the provision of legal rights, such as free- countries of the developing and developed world. In Central ty distribution.19 For example, in Buenos Aires, Argentina, a Africa, Asia and Latin America, representing more than 70 per
hold or long-term leases, if these are possible and desirable. Europe, for instance, more than half the inhabitants rent the study found that the allocation of property rights across slum cent of a global sample of cities, acknowledged that they did
In formal, advanced systems, tenure rights are reflected in laws houses in which they live. Yet, their tenure is extremely secure. households is usually not random, but based on wealth, family not know to which extent urban growth could be attributed to
and regulations governing housing and property rights. In devel- And despite a significant increase in ownership in North characteristics, political patronage and other mechanisms that informal settlements; 40 cities, or 20 per cent of the sample,
oped countries, security of tenure is in most cases guaranteed, America that saw home ownership rise - from 64 per cent to 69 mark differences between those who have property rights and provided some general data as a percentage of the total urban
and people enjoy a higher order of rights that enables them to per cent from 1993 to 2003 - a third of the region’s inhabitants those who do not.20 Slum upgrading projects in other countries growth; less than 20 cities, or 10 per cent of the sample, made
sell, rent, improve, develop, sub-let or inherit land or property. In still do not own their own home.14 Even in the developing have also been known to play into the hands of illegal structure available accurate information in square kilometers, as request-
virtually all developed countries, these rights are further embed- world, studies have shown that ownership is neither necessary owners and negligent landlords, who lay claim on upgraded ed by the Programme.26
ded in infrastructure, land administration and land recording nor sufficient to generate tenure security. A Demographic and dwellings in order to extract more rent from tenants or to sell Consequently, the number of people lacking secure tenure is
mechanisms. Rights derived from land and property are defined Heath Survey conducted in Senegal in 2005 shows that 45 per them off to higher-income groups.21 Upgrading policies based not known in most cities and countries of the developing
in a way that makes them easy to identify in terms of boundaries, cent of the inhabitants owned their homes, while 42 per cent on ownership and large scale granting of individual titles are world. The lack of official data on informal growth is sympto-
demarcations, registration and transactions. Individuals (owners were tenants. Only 40 per cent of the so-called owners claimed also extremely expensive and cumbersome, especially in coun- matic of the poor capacity of local authorities to plan the
and tenants) have a clear understanding of the potential that land to have title deeds, while 25 per cent had a certificate of occu- tries where titling systems are slow, laborious, inflexible and urbanization process. Instead of learning to accept inevitable
and housing offers in terms of use, appropriation and trade. UN- pation or a receipt of purchase. A significant proportion – 13 generally unaffordable – not to mention prone to corrupt prac- urban growth in informal settlements and slums, governments,
HABITAT uses the concept of “tenure advantage” to describe per cent – had no formal or informal authorization to occupy tices that harm rather than benefit the urban poor. In the like the proverbial ostrich, have chosen to bury their heads in
these more advanced rights of individuals and households.13 the dwelling. Of the tenants, only 14 per cent had a formal Philippines, for instance, establishing legal ownership takes 168 the sand, hoping the problem will go away. That is why meth-
However, it is understood that slum dwellers can also sell, rent or contract and a staggering 68 per cent did not have any kind of procedures and between 13 and 15 years.22 All these problems ods to capture and measure informal settlement growth are not
improve the land or property they occupy, but in a less secure document to prove tenancy. Yet 76 per cent of both owners and are compounded by the fact that little is known about the perceived as necessary. It is also the reason why the extent and
environment. Tenure advantage rights are also known as “trans- tenants said they enjoyed security of tenure.15 severity or range of insecure tenure within cities of the develop- scope of tenure insecurity is not known in most cities and
ferability rights” that have a direct bearing on the livelihoods of The view that ownership is the only path to security has ing world, which makes it difficult to make appropriate inter- countries of the developing world.
people, as they are extensively used as security from which capi- gained credence among development agencies and practi- ventions at the policy level or in the implementation of upgrad- There have, however, been exceptions to this rule. Some gov-
tal can be derived. As property markets become more active, tioners who argue that urban poverty can be drastically ing and regularization programmes. ernments have addressed urban tenure security in their plans,
transferability operations increase because public and private reduced if slum dwellers acquire ownership rights that can In developing countries, customary ownership, religious sys- providing land to urban dwellers before occupation. Others
infrastructure supports the activities. Land is extensively used as enable them to secure loans to improve their housing and to tems governing land issues and informal agreements between have integrated tenure security in their housing programmes

96 97
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

ing to establish monitoring systems on issues such as evictions

and other housing rights abuses, even they admit that a com-
prehensive eviction-monitoring system remains elusive as many
gaps in coverage remain because for every reported eviction,
there is an unknown number of unreported cases.28 In the
absence of a monitoring framework that provides a reference
point and guidance, it is statistically difficult to prove whether
tenure has improved or deteriorated, as evidence remains most-
ly anecdotal or based on media reports.
As long as mainstream systems of data collection and analy-
sis (censuses and surveys) do not recognize secure tenure as a
unit of analysis, data will not be periodically produced.
El Alto, Bolivia ©RUSSELL PEARCE. IMAGE FROM BIGSTOCKPHOTO.COM
Endnotes
1 Much of the data on evictions in this chapter is drawn from various 31 UN-HABITAT currently undertakes this monitoring strategy in collabora-
reports by the Centre on Housing Rights and Evictions (COHRE). tion with several partners, namely DFID, the World Bank, and the gov-
2 COHRE 2003. ernments of Canada, USA and Belgium.
3 COHRE 2004.
4 The most comprehensive interpretation of the scope of protection
against forced evictions was made by the UN Committee on Economic,
Social and Cultural Rights in its general comment No 7, adopted in
1997, while the right to adequate housing is enshrined in many interna-
tional human rights instruments, most notably the Universal

Informal growth has become the most
significant mode of housing production in
cities of the developing world.
and projects, but in a very sporadic way. Many governments
have responded to informal occupations through remedial
actions of regularization; however, their interventions have nei-
ther been systematic nor politically disinterested. The majority
of the governments have opted to ignore informal settlements
altogether, either because of their unwillingness to accept in-
migration and urban growth, or their incapacity to cope with
the accelerated process of urbanization.
The problem is compounded by the fact that while United
Nations and other agencies have been testing and developing
systems of monitoring global poverty, disease, illiteracy, unem-
ployment, and other indicators over the past five decades, the
operationalization of the secure tenure concept, as part of a
global monitoring system, remains challenging. Indeed, at the
present time, it is neither possible to obtain household-level
data on secure tenure, nor to produce global comparative data
on various institutional aspects of secure tenure.27 Although a
growing global network of organizations, such as the Centre on
Housing Rights and Evictions (COHRE), Amnesty
International and Human Rights Watch, among others, are try-
ECLAC, for instance, recognizes the difficulties in determining Declaration of Human Rights and the International Covenant on
secure tenure in Latin America and the Caribbean and can only Economic, Social and Cultural Rights.
estimate the level of informality in the region’s the cities, which 5 United Nations 2005c.
it places at between 10 and 15 per cent of the urban population 6 These “apartheid cities” appear to be more prevalent in former British
in Argentina and Uruguay, between 20 and 40 per cent in colonies, such Kenya and Zimbabwe, where the migration of the indige-
Mexico and Peru and between 50 and 70 per cent in Ecuador nous population to cities was highly regulated and often prohibited dur-
and Honduras.29 The World Bank estimates that more than 50 ing the colonial period.
per cent of the peri-urban population in Africa and more than 7 World Bank 2003.
50 per cent in Asia has some form of informal tenure. But these 8 De Soto 2000.
are at best estimates as few countries produce data on secure 9 Fernandes, et al. 1998.
tenure, which can be used as basis for global and regional mon- 10 Inter-American Development Bank 2004.
itoring or to assess progress or setbacks. 11 Government of Indonesia 2005.
12 Demographic and Health Survey, Nicaragua, 2001.
13 Bazoglu & Moreno 2005.
■ Toward a Global Monitoring Strategy on 14 US Census Bureau, American Housing Survey (1993 and 2003).
Secure Tenure 15 UN-HABITAT 2005c.
16 De Soto makes a strong argument for ownership in The Mystery of
In the last thirty years, security of tenure has been part of the Capital (2000).
conceptual, institutional and technical discussions about land 17 Lee & Ghanime 2004.
and housing policies. In some moments of this saga, tenure 18 Economic Commission for Latin America and the Caribbean 2005.
security has received a great deal of attention, particularly dur- 19 Quan, et al. 2005.
ing international conferences, political declarations and the 20 Goytia 2005.
preparation of technical reports. These discussions, however, 21 Huchzermeyer 2006.
have not resulted in greater efforts to integrate tenure security 22 UN-HABITAT 2004c.
in policy reforms and urban interventions. In fact, no mecha- 23 Augustinas 2003.
nism currently exists to monitor secure tenure as part of 24 World Bank 2003.
Millennium Development Goal 7, target 11 on improving the 25 UN-HABITAT 2004d.
lives of slum dwellers.30 26 UN-HABITAT training workshops in five regions of the developing
UN-HABITAT and its partners are working on the prepara- world. Urban Indicators Programme, Cluster B Urban Data, Nairobi,
tion of a global monitoring system31 that could in the future 2005.
provide a framework to assist governments at local and nation- 27 Bazoglu & Moreno 2005.
al levels to produce estimates at the household levels on how 28 COHRE 2004.
many people have secure tenure, using a consistent methodol- 29 Clichevsky 2003.
ogy in terms of definitions, indicators and variables. The mon- 30 A list of 18 targets and more than 40 indicators corresponding to these
itoring system would serve to track changes in land and resi- goals ensure a common assessment and appreciation of the status of
dential secure tenure to measure how the right to adequate the Millennium Development Goals at the global, national and local lev-
housing is progressively realized and how slum dwellers are els. Among the indicators for monitoring progress on the Goals, secure
improving their living conditions. It would also be an advoca- tenure was given a prominent place (indicator 32, “the proportion of
cy and policy instrument to bring together policy formulation, households with access to secure tenure”), as part of Target 11 “by
action and monitoring activities; otherwise, policy actions will 2020, to have achieved a significant improvement in the lives of at
continue to be formulated independently of results, without least 100 million slum dwellers” of Goal 7 “Ensure Environmental
clearly indicating if there is efficient, equitable and sustainable Sustainability”.
progress in attaining target 11.

98 99
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
One of the affected areas in Harare prior to the May 2005 evictions.
Source: IKONOS: Copyright INTA Space Turk 2005; QUICK BIRD: Copyright Digital Globe 2005; Image processing and analysis: UNOSAT.
A victim of eviction in Zimbabwe UN-HABITAT
Demolition of a backyard extension.
100
The same area after the May 2005 evictions.
A victim of eviction in Zimbabwe UN-HABITAT
Evicted family.
Evictions Fail to Address the Root Cause of Urban Poverty in Zimbabwe
In May 2005, with little or no warning, the Government of
Zimbabwe embarked on an operation to “clean up” its cities.
“Operation Murambatsvina”, or Operation Restore Order, started in
the capital Harare, and rapidly evolved into a nationwide demoli-
tion and eviction campaign carried out by the police and the army.
Popularly known as “Operation Tsunami” because of its speed and
ferocity, it resulted in the destruction of homes, business premises
and vending sites in several parts of the country. A July 2005 report
by the UN Special Envoy on Human Settlements Issues in
Zimbabwe, Mrs. Anna Tibaijuka, estimated that some 700,000 peo-
ple in cities across the country had either lost their homes, their
source of livelihood or both as a result of the Operation and a fur-
ther 2.4 million were indirectly affected in varying degrees.
Operation Restore Order took place at a time of persistent budget
deficits, triple-digit inflation, critical food and fuel shortages and
chronic shortages of foreign currency. It was implemented in a
highly polarized climate characterized by mistrust, fear and a lack
of dialogue between the government and local authorities, and
between the government and civil society. Although the econom-
ic crisis was precipitated by a variety of factors,including increasing
isolation by Western powers, many of Zimbabwe’s problems pre-
cede the country’s independence.
Zimbabwe achieved independence in 1980 amid promises of peace
and prosperity. While the government successfully provided social
services,such as education and health care,and increased wages for
the black majority during the early years of independence, underly-
ing socio-political and economic problems were left unresolved
and eventually produced a national crisis. Of these, the land ques-
tion was the most problematic.While the liberation war was fought
over land, historical inequity was embedded in the constitutional
settlement agreed upon at independence that preserved colonial
patterns of land ownership.To make matters worse, a failed attempt
at structural adjustment in the 1990s led to massive retrenchment
of civil servants, closure of manufacturing industries, inflation and
deterioration of basic services. In February 1998, peasants took mat-
ters into their own hands by staging illegal – and politically motivat-
ed – invasions of commercial farms, forcing the government to initi-
ate a “fast-track land reform programme” in 2000.
It is against this background that Operation Restore Order took
place. Ironically, while the government tried to appease the coun-
try’s rural population, it took a rather elitist approach with its urban
citizenry by imposing stringent by-laws and standards that
deemed many dwellings in the city “illegal”. As the report of the UN
Special Envoy states,“The nationalist elite seemed to have perpet-
uated the colonial mentality of high standards for a few at the
expense of the majority. In the end, while the liberation struggle
was against the ‘white settlers’ and the economic and political
power they monopolized, the government was not able to reverse
the unequal and exploitative nature of colonial capitalism itself.”
Sources: United Nations 2005c; UN-HABITAT 2003a; IRIN News 2005.
Like many former British colonies in sub-Saharan Africa, urban
planning in the country formerly known as Rhodesia typically
reserved the city core for whites, while leaving an undeveloped
buffer space around the central business district.Towns were often
pre-planned and imposed on localities, without much attention
being given to existing constraints.The indigenous population was
either relocated to black townships on the outskirts of the city or to
rural “reserves” to make room for European settlers.
The indigenous African population moved to towns and cities in
large numbers only after attaining independence when policies
prohibiting their movement to cities were abolished or discarded.
This resulted in a major shift of populations from rural to urban
areas. Within a decade of independence, Zimbabwe’s urban popu-
lation rose from 23 per cent in the 1980s to 30 per cent by the early
1990s. However, stringent by-laws and standards adopted from the
colonial administration ensured that Zimbabwe’s cities remained
largely immune to the explosive growth of slums and squatter set-
tlements that are characteristic of other African cities.Official statis-
tics compiled by UN-HABITAT show that in 2001 only 3.4 per cent
of the urban population in the country lived in slums,a figure much
lower than that of even industrialized countries that had about 6.2
per cent of their population living in slum-like conditions, and dra-
matically lower than that of other African cities, where between 30
per cent to 70 per cent of the urban population lives in slums.
The acquisition of peri-urban farms during the fast-track reform
programme in 2000 provided one of the first opportunities for the
urban poor to occupy land in the vicinity of the city, many of which
were in the form of “backyard extensions” of legal dwellings.These
extensions provided affordable rental housing to the city’s poor
and were a source of much-needed income for the owners. Most of
these extensions within the cities have now been demolished,
affecting hundreds of thousands of women, men and children who
are sinking deeper into poverty and rendered more vulnerable. A
follow-up report by the UN Office for the Coordination of
Humanitarian Affairs (OCHA), for instance, found that some of the
worst affected were women and children living with HIV/AIDS.
(Zimbabwe has one of the world’s highest HIV prevalence at about
a quarter of the total population.) A survey by the Bulawayo-based
Matabeland AIDS Council, for instance, found that many of those
displaced by the “clean-up”campaign could not continue with their
treatment and were in dire need of drugs.
It is for these reasons that the UN Special Envoy’s report recom-
mends, among other things, that outdated laws be suspended or
reviewed in order to align them to the social, economic and cultur-
al realities facing the majority of the country’s population, namely
the poor. It also recommends that the international community
draw lessons from the Zimbabwe crisis for the entire continent of
Africa by ensuring that policies aimed at reducing poverty do not
have the opposite effect.
101
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

3
Part Three

Where We Live Matters

This Part provides concrete evidence of how

inadequate housing and lack of basic services
threaten the health, education and employment
opportunities of slum dwellers. Using data that
goes beyond the conventional urban–rural
dichotomy, the Part presents, for the first time,
disaggregated information at slum and non-slum
levels that help us to understand the connection
between living conditions and human development.
This connection is fundamental to appreciate both
the vulnerability of slum dwellers and the levels of
poverty and deprivation that they experience,
particularly in relation to social and health
outcomes.
Mexico City, Mexico TERRAZAS GLAVAN MONICA/UNEP/STILL PICTURES
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
3.1 The Social and Health Costs of Living in a Slum
The Urban Rural Divide ©FEDOR SIDOROV. IMAGE FROM BIGSTOCKPHOTO.COM
How do inadequate water supply and
overcrowding in slums impact child mortality
rates? Is urban insecurity related to
inequality within cities? How do conflicts in
rural areas exacerbate slum formation in
urban areas? Why are women who live in
slums more likely to be infected with HIV
than their rural counterparts? Are the
Millennium Development Goals being met in
the slums of the world? How does one’s
physical address influence one’s health,
education and employment opportunities?
Does it matter where we live?
104
A
nyone who has dealt with real estate agents
knows the mantra “location, location, location”:
place and progress are inextricably intertwined.
This is especially so for the world’s urban poor.
Indeed, as the following chapters illustrate,
where we live can have a significant influence on whether or
not we are likely to be healthy, educated, employed, safe, or
impoverished. UN-HABITAT analyses of recent survey data
show that people who live in slums face serious threats to their
well-being. In some cases, living in a crowded, unsanitary slum
is even more life-threatening than living in an impoverished vil-
lage. Some studies have also shown that job applicants from
slum communities are less likely to be interviewed than those
living on “the right side of town”. In other words, living in a
slum often means being more vulnerable to a host of social and
economic threats that make the achievement of the
Millennium Development Goals in cities both a major chal-
lenge and an urgent need.
The Millennium Development Goals have been accepted
internationally as a common development framework. At their
core, the Goals aim to bring the vast majority of the world’s
population out of a poverty trap that robs them of their health,
dignity and aspirations for fulfilling their human potential.
UN-HABITAT has been assigned the responsibility of assisting
Member States of the United Nations to monitor and attain
Millennium Development Goal 7, target 11: by 2020, to have
achieved a significant improvement in the lives of at least 100 mil-
lion slum dwellers.
The inclusion of the slum target in the Millennium
Development Goals indicates a recognition by the internation-
al community that urban poverty is a growing challenge.
However, national and international data and poverty reduc-
tion strategies still do not acknowledge the deprivation levels in
slums and consistently underreport health, literacy and other
development indicators. For instance, while aggregate health
statistics suggest that urban dwellers have better health status
compared to those living in rural areas, UN-HABITAT indica-
tors show that there is a large and growing gap between the
health status of high-income urban residents and those living in
poverty at the margins of society.
The internationally agreed-upon slum target has been large-
ly ignored in country and agency reports on progress on the
Millennium Development Goals, due in part to the lack of
intra-city data disaggregated across slum and non-slum areas. A
review of the existing strategies to improve in the lives of slum
dwellers reveals a gap in addressing the situation of the urban
poor in national and international programmes. Most national
reports underestimate the level of urban poverty; moreover, the
measurement of poverty in both rural and urban areas is based
on income, which often does not provide an accurate picture of
the scale and multidimensional nature of poverty experienced
by the urban poor. The crisis that slum dwellers are facing has
been masked by the common practice in social science to
analyse the human settlements dimension by categorising
information according to “urban” and “rural”. In country
reports, all urban households – rich and poor – are averaged
together to provide single estimates of poverty, education,
health, employment, and human settlements, leading to an
underestimation of the urban poor and the conditions in which
they live.1 Another aspect that gets lost in urban averages is
intra-city inequality. Studies show that the decreased mortality
recorded in urban areas in the 1990s was primarily a result of
high-income residents living longer, indicating widening health
disparities between the rich and the poor. This trend is partic-
ularly prevalent in Latin America.
Data produced by the United Nations, World Bank and
other agencies presents urban poverty on a regional scale and
generally links it to theoretical projections not based on actual
surveys. For instance, World Bank projections indicate that the
locus of poverty will move to cities only after 2035.2 This pro-
jection serves as an “early warning system”, much like the warn-
ings issued by the World Health Organization (WHO) in the
1980s about the impending AIDS crisis. However, other agen-
cies have a more realistic view of the scale of urban poverty and
believe that the crisis has already begun.
As discussed elsewhere in this Report, although the number
of slum dwellers is not an accurate measurement of the num-
ber of urban poor – poverty can manifest itself in non-slum
areas; conversely, not all people who live in slums are poor –
slums are a physical dimension of urban poverty. It is, there-
fore, crucial to know how many slum dwellers there are, where
they are located, and what their basic needs are in terms of shel-
ter, water, sanitation, health, education, employment, and the
like. Part One of this Report highlights the numbers and loca-
tions of slum dwellers around the world; this section addresses
their needs in terms of specific Millennium Development Goal
targets and indicators.
UN-HABITAT’s analyses of disaggregated urban data point
to some key findings. Child mortality rates in poor urban and
rural communities, for instance, are much higher than those of
high-income urban communities. Furthermore, families living
under conditions of severe shelter deprivation experience, in
some countries, a child mortality rate three times higher than
that of families that enjoy full use of safe water, improved san-
itation, durable housing, and decent living conditions. The fact
that inequalities based on socio-economic disparities are so per-
sistent in urban areas of developing countries implies that
reliance on global average statistics to allocate resources
between rural and urban areas could be dangerously mislead-
ing. Lack of basic shelter services, as a correlate of poverty, is the
expression of various social and health issues such as low edu-
cation, wide gender inequalities, poor maternal and child
health, and hunger. Poor living conditions also contribute to a
host of diseases and infections, such as diarrhoea, acute respira-
tory infections, malaria and HIV/AIDS. In terms of education,
studies indicate that a majority of parents settling in slums
postpone sending their children, especially girls, to school,
until they are able to manage other expenses, such as food, rent
and transport.
In this Report, UN-HABITAT aims to show that improve-
ment in the lives of slum dwellers leads to progress on the
achievement of all of the Millennium Development Goals. By
improving slums – or preventing their formation – govern-
ments are also eradicating poverty and hunger, increasing liter-
acy, combating HIV/AIDS, reducing child mortality, improv-
ing the environment, and promoting gender equality. This calls
for the localization of the Goals: local policy needs to be
informed about the consequences of persistent inequalities in
cities and the myriad problems associated with the living con-
ditions of poor urban communities.
Endnotes
1 Fry, et al. 2002.
2 Ravallion 2001.
105
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
3.2 Hunger: The Invisible Crisis in Cities
E
radicating extreme poverty and
hunger is the first Millennium
Development Goal. Reducing
the proportion of people suffer-
ing from hunger in the world is
therefore acknowledged as essential to achiev-
ing all of the other Goals. The United Nations
Millennium Development Goals Report 2005
states that there were 815 million hungry peo-
ple in the developing world in 2002 and that
“most of the world’s hungry live in rural areas
and depend on the consumption and sale of
natural products for both their income and
their food”. The report adds that sub-Saharan
Kibera, Nairobi HIROSHI SATO
Africa and Southern Asia are the worst-affect-
ed regions and that “hunger tends to be con-
centrated among the landless or among farm-
ers whose plots are too small to provide for
their needs”.
Although hunger is most often associated
with low agricultural output, drought and
famine in rural areas, various studies have shown that hunger is
not always related to food production or availability; rather, in
urban areas, other factors, such as low incomes, inadequate
access to basic services and poor living conditions, play more
significant roles.
■ What makes hunger in cities unique?
In rural communities, exogenous factors such as geography
and climate are major determinants of food availability and
dietary intake. Rice is generally consumed in the humid trop-
ics, while millet is more frequently eaten in arid regions. People
who live in mountainous areas are limited to barley and pota-
toes at the highest altitudes, but a variety of cereals can be pro-
duced in lowland valleys. Pastoralists are more likely to rely on
their animals for food. Unless sophisticated market systems
have been developed, the inhabitants of a particular ecological
zone consume only what they can produce locally. There may
also be dramatic differences in the types and amounts of food
available in different seasons.1
Whereas crop patterns, size of land and the time and quality
of the harvest often determine food availability for the family of
the subsistence farmer, disposable income and food prices large-
106
ly determine the amount and types of foods consumed by low-
income families in urban areas.2 In cities, hunger is usually the
consequence of people’s inability to purchase food that is both
sufficient and nutritious. An assessment of the “food basket” of
slum households shows that it is mainly composed of items low
in calories and vitamins,3 making these households more prone
to malnutrition.
Even in situations where a country produces enough food to
feed everyone, hunger may persist in urban areas. In fact, the
situation of the urban poor can be worse during famines and
droughts than the situation of villagers; international food aid
distributed during difficult times is concentrated in rural
areas, while in cities, prices for essential food products pro-
duced within the country soar during such times, adversely
impacting the ability of low-income people to purchase food.
When inflation hits food supplies, poor urban families may be
forced to use up to 70 or 80 per cent of their disposable
income to purchase food, which often means that they have
little money left over to pay for non-food items, such as rent,
school fees and transport. Thus, variations in income or food
prices directly translate into rising rates of malnutrition in
urban areas. In poor urban communities, even seasonal vari-
ability in income or food availability can lead to seasonal
swings in malnutrition.
©BARBARA WHITE. IMAGE FROM BIGSTOCKPHOTO.COM
Homeless woman in Chicago
When inflation hits food supplies, poor
urban families may be forced to use up
to 70 or 80 per cent of their disposable
income to purchase food.
■ The link between inadequate shelter and
hunger
UN-HABITAT analyses indicate that hunger and malnutri-
tion is particularly high in slums and in rural areas, where access
to adequate housing and basic services, such as safe water and
sanitation, is poor or non-existent. The poor living conditions
prevalent in slums and in rural areas impact people’s ability to
avert hunger and malnutrition in various ways. Households’
sources of drinking water and methods of waste disposal impact
children’s nutritional status, as diarrhoea and other diseases
resulting from inadequate water and sanitation can prevent
young children from absorbing nutrients and growing properly.
Without an adequate and safe supply of water, a household’s
personal, domestic and food hygiene are compromised and the
risk of contamination and diseases – including diarrhoea and
acute respiratory infections – increases. Overcrowded slum
households are also more likely to use inadequate sanitation and
to share toilet facilities with many other households, which
increases the risk of diarrhoea and respiratory infections. At the
community level, lack of waste management and wastewater
treatment increases the prevalence of diseases such as diarrhoea,
acute respiratory infections and malaria, all of which impact
nutritional status and overall health.
When levels of child malnutrition are used to measure
hunger, evidence suggests a strong link between malnutrition
and slums. That is, places that report a high prevalence of
child malnutrition typically have correspondingly high levels
of slum incidence.4 For instance, countries such as
Bangladesh, Ethiopia, Guatemala, Haiti, India, Nepal and
Niger – all of which have a high incidence of slums – are also
those with among the highest prevalence of malnourished chil-
dren. In these countries, 4 out of 10 children in slums are mal-
nourished – a proportion 20 times higher than that of developed
countries. In some countries, incidence of child malnutrition in
slums is almost the same as that of rural areas. In Ethiopia, for
instance, child malnutrition in slums and in rural areas is 47 per
cent and 49 per cent, respectively, compared with 27 per cent in
non-slum urban areas. Similar findings are reflected in Niger,
where child malnutrition is 50 per cent in slums, 52 per cent in
rural areas and 35 per cent in non-slum urban areas.
In general, malnutrition is much higher in rural and slum
areas than in non-slum urban areas, even in countries with low
levels of slum incidence. In Morocco, slum and rural children
are twice as likely to be malnourished as their non-slum coun-
terparts; while 7 per cent of children in non-slum areas are mal-
nourished, 14 per cent of children in both slum and rural areas
are malnourished. The greatest inequalities exist in Brazil and
Côte d’Ivoire, where child malnutrition is three to four times
higher in slums than in non-slum areas (19 per cent versus 5 per
cent, and 37 per cent versus 10 per cent, respectively).
The relationship between malnutrition and poor living con-
ditions is illustrated by various studies that show that malnutri-
tion levels decrease when investments are made to improve serv-
ices and infrastructure in low-income areas. The greatest decline
in malnutrition in Eastern Asia, for instance, happened when
China significantly improved its food distribution networks and
health facilities, and provided increased access to improved
drinking water. A study conducted in India in the 1950s attrib-
uted much of that country’s rise in life expectancy (from about
25 years to 50 years) in the first half of the 20th century to the
prevention of recurrent famines that had characterized the sub-
continent’s history; this was achieved by stabilizing food sup-
plies with railroads, road networks, irrigation, food distribution
markets and political security.5
■ The urban penalty
There is increasing evidence of what UN-HABITAT refers to
as the “urban penalty”: a number of key health indicators for vul-
nerable urban populations are as bad as or worse than those of
rural populations. Despite the improved coverage of health serv-
ices and basic service delivery in some countries, certain popula-
tion groups have been left behind and opportunities remain
unevenly distributed. This is particularly true in slum settlements
around the world, which are as disadvantaged as rural popula-
tions, especially in least-developed countries with high urban
growth rates.
Hunger eradication strategies must embrace multiple interven-
tions, not only those related to food availability, but also those
related to shelter. Access to adequate housing, safe water and ade-
quate sanitation do improve the nutritional status of slum dwellers
and rural populations, with or without an increase in food avail-
ability. This justifies a comprehensive approach that includes
strong linkages between slum upgrading and the sustainability of
programmes delivering health and nutrition services.
107
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

Endnotes

1 Bidinger, et al. 1986.

Agrocities: Combating Hunger in Urban Areas
2 Ibid.
In many cities, particularly in the devel- Zimbabwe’s capital city, Harare, was val- urban poor families and identifies vacant
3 See Demographic and Health Survey comparative report, ORC Macro
oping world, urban agriculture has ued at $25 million and covered more public and private spaces that could be
2004.
helped increase food security and than 9,000 hectares. used to grow chemical-free vegetables,
4 For a better understanding of hunger and food deprivation in cities,
reduce hunger among vulnerable popu- but has also established a food produc-
UN-HABITAT has analysed Demographic and Health Surveys and
lations. Evidence suggests that urban Less is known, however, about the quali- tion system that practises low-input
Multiple Indicator Cluster Surveys data on child nutrition in Africa,
agriculture – the practice of growing, ty and safety of food produced in urban agriculture using appropriate technolo-
Asia and Latin America and the Caribbean. Child malnutrition is
raising, processing and distributing food or peri-urban areas where soil and water gies that are easy for the urban poor to
assessed by the proportion of children underweight, and associated
in and around an urban area – con- contamination levels are higher than in adopt. It has also ensured that the veg-
variables. Underweight, defined as low weight for age, takes into
tributes significantly to urban food sup- rural areas. In cases where quality has etables produced are sold in strategic
account both acute malnutrition (wasting) and chronic malnutrition
ply and household food security, particu- been tested, the results have been markets within the city and that urban
(stunting). A child can be underweight for his or her age because he
larly among low-income groups. mixed. Findings for fish in Calcutta and agriculture is institutionalized as a local
or she is has suffered from “wasting”, “stunting” or both (UNICEF
for vegetables in Accra, for instance, government policy that is incorporated
2003). Wasting may be the result of inadequate food intake or recent
Surveys conducted in the late 1990s by a showed no difference in contamination in the city’s strategic plan.
episodes of illness causing loss of weight and the onset of malnutri-
range of institutions in 24 cities, mainly in levels between urban-based and rural-
tion. Among adults, this is defined as food consumption insufficient to
Africa and Asia, and urban areas of based supplies but produce from indus- Since the programme started in 2001,
meet minimum levels of dietary energy requirements. Stunting
Bulgaria, Romania and the Russian trial sites in Poland showed higher levels 791 urban kitchen gardens have been
reflects failure to receive adequate nutrition over a long period and
Federation, showed that households of heavy metals. A recent newspaper set up, providing employment to over
may also be caused by recurrent and chronic illness; it represents a
involved in growing some of their food report in Kenya also warned that an 5,000 families. An additional 10,000 fami-
measure of the long-term effects of malnutrition in a population (Food
made up anything from an important increasing number of “urban farmers” in lies are directly linked to the production
and Agriculture Organization of the United Nations 2002). For global
minority to a large majority of all house- and around the capital of Nairobi were of chemical-free vegetables that feed
Fruit vendor in Kabul RASNA WARAH

trends, malnutrition is assessed for both adults and children.

holds in any given city. The surveys also
However, due to data limitations, UN-HABITAT intra-city differential
showed that poor households that prac-
analyses focus on children, and to some extent mothers, while recog-
tised urban agriculture ate more meals
nizing that hunger also affects adults.
and had more balanced diets than those
5 Davis 1951.
households that did not rely on urban
agriculture for their food supply. Self-pro-
visioning also helped urban poor house-
holds to save money that they might
previously have used to purchase food.
In many cities, urban agriculture is a
FIGURE 3.2.1 SLUM INCIDENCE AND PROPORTION OF UNDERWEIGHT CHILDREN IN SELECTED COUNTRIES
main or supplementary source of
income or employment among low-
income households. Generally, the high-
er the market value of the produce, the
In general, malnutrition is much greater its contribution to household
50 higher in rural and slum areas income. A survey conducted in Lome,
than in non-slum urban areas Togo, for instance, showed that market
40 gardeners earned 10 times the monthly
Percentage of under weight children
vandalising sewage networks to fertilize
their vegetables, a practice that is affect-
ing the quality of vegetables produced.
Experts say that since sewage networks
often receive non-domestic waste, they
could contain high levels of heavy met-
als and organic matter that does not
decompose easily, which could seriously
affect the health of consumers of
sewage-fed vegetables.
Rosario’s kitchen gardens
Urban agriculture is not only an impor-
tant source of food and income among
urban poor households – it can also sup-
port a wide range of economic activities
related to the production, sale, marketing
and consumption of produce. In Rosario
over 40,000 people in this city with a
population of 1.3 million. The pro-
gramme has also led to the creation of a
network of 350 groups that participate
in local fairs where the produce is mar-
keted and sold.
The formal recognition of urban agricul-
ture as a legitimate urban land use poli-
cy has enabled the municipality to set
up a register and a Geographical
Information Systems (GIS) databank on
potentially productive land areas in the
city. It has also improved the tenure sta-
tus of the urban poor through user-
rights agreements and tax incentives for
land owners who make their land avail-
able for urban agriculture. The demand
for the vegetables has also increased as

minimum wage. in the province of Santa Fe in Argentina, the local fairs are the only places where
30 for instance, the Urban Agriculture residents can access organic produce.
Moreover, the quantity of food supplied Programme has been implementing an
20
through urban agriculture comprises a urban kitchen gardens project that has For the urban poor, the kitchen gardens
Rural significant amount of the total food con- helped improve food security in the city, are not only an important source of
Slum sumed in cities and is worth tens of mil- generated income for urban poor fami- employment but also a source of nutri-
10 Non-slum
lions of dollars. In the late 1990s, milk lies and transformed uncultivated land tious, chemical-free food. As a result of
produced each year in Dar es Salaam, into productive spaces. the success of the programme, proposals
0 Tanzania, was estimated to be worth are being developed to incorporate
ia

ia

ia

da

sh

il

more than $10 million. In the mid-1990s, The Urban Agriculture Programme of urban agriculture in future settlement
ke

az
bi

di

bi

gu

al
op

ib

de
an
za

In

em
Br
r
m

ra
hi

Tu
Za

la

lo
Ug
n
Na

ca
Et

at
Ta

ng

Co

rain-fed maize in and around Rosario not only provides training to and housing plans.
Gu
Ni
Ba

Sources: Mougeot 2005a & 2005b; www.bestpractices.org; Musa 2005.

UN-HABITAT, 2005 Urban Indicators Programme, Phase III.
Source: Demographic and Health Surveys 1995-2003.

108 109
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

3.3 The Urban Poor Die Young

FIGURE 3.3.1 UNDER-FIVE MORTALITY (DEATHS PER 1000 BIRTHS) BY TYPE OF RESIDENCE IN SELECTED COUNTRIES

250

200

Deaths per 1000 births

150

100

Urban
Santos Praia, Brazil TOPHAM PICTUREPOINT

Rural
50
Non-slum
Slum

ria

ric h

ia

da

il

sh

an

am
yp

az

ne
cc

ni

oo

Af out

bi

al

di

ta
an

de
an

st
ge

a
Eg

Be

em

In

N
kis
Br

pi
o

er

nz

kh
S

lo

la
or

Ug
Ni

et
ilip
m

be
at
Ta

Co

ng

za
M

Vi
Ca

Gu

Ph

Uz
Ka
Ba
Source: UN-HABITAT 2005c.
Note: Computed from Demographic and Health Surveys (DHS) data 1995-2003.

C
hild mortality is closely linked to poverty, and
child mortality rates are reliable indicators of
human and economic development in coun-
tries. Millennium Development Goal 4 aims to
reduce child mortality by cutting the worldwide
under-five mortality rate by two thirds between 1990 and
2015. Doing so will require a special focus on the most vulner-
able young children and families – those living in rural areas
and in urban slums. Inadequate shelter and poor living condi-
tions in slums are related to a host of health risks, including
exposure to infectious diseases and indoor air pollution that
shorten the life span of slum dwellers. This chapter describes
the major health risks for slum dwellers and argues that even
simple improvements in their living conditions can save lives.
■ Intra-city disparities in child mortality
Mortality rates often reflect
inequalities in access to shelter,
health care and education.
Five diseases – pneumonia, diarrhoea, malaria, measles, and
HIV/AIDS – account for more than 50 per cent of all child
deaths. The chances of contracting any one or a combination
of these diseases are compounded by poor living conditions
and poor access to health services. Mortality rates often reflect
The ratio of child deaths in slum areas to child deaths in non- Where child mortality rates are high,
slum areas is consistently high in all developing countries, even in the proportion of slum households is
countries that have made progress toward reducing child mortal-
ity overall. Several studies show that mortality differentials across
typically also high.
groups tend to narrow only if policies focus explicitly on increas-
ing equity in access to healthcare and safe housing. Without such
a focus, improvements in the average rate may not reflect real
improvements for both disadvantaged and advantaged socio-eco-
nomic groups.2 In other words, only when governments develop
health policies that address the needs of the most vulnerable pop- FIGURE 3.3.2 UNDER-FIVE MORTALITY (DEATHS PER 1000 BIRTHS)
ulations do child mortality rates decline. BY TYPE OF RESIDENCE IN SELECTED CITIES
Much evidence supports the theory that inequality breeds ill
100
health. The World Health Organization (WHO) concedes that
“being excluded from the life of society and treated as less than 80
equal leads to worse health and greater risks of premature death.

Deaths per 1000 births

Child mortality rates in developing countries are 10 times
higher than those in the developed world. In 2003, sub-
Saharan Africa and Southern Asia had the largest share of chil-
inequalities in access to shelter, health care, employment, and
education among different socio-economic groups. High
mortality rates in slums are also compounded by the fact that
millions of slum dwellers live on hazardous sites that are
60
The stresses of living in poverty are particularly harmful to [preg-
nant women], babies, children and old people.”3 Medical research 40
confirms the fact that income distribution is a more powerful
determinant of health and mortality than the overall wealth of 20

dren who died before reaching their fifth birthday.1 Child mor- prone to natural disasters, such as floods, or that are located nations.4 The lessons learned from the public health experience of
0
tality rates appear to be closely related to urban poverty levels, in or near toxic areas, such as garbage dumps, quarries or fac- developed countries point to some patterns. In the United

ca

an

iro

a
ba

ka

ol

ib
and particularly to the incidence of slum households, as Kingdom, differences in the health experiences of various groups

an

rb

ne
tories. Children are particularly at risk of illness and death as

To

Pa

rit
Ra

Da

Du
bl

Ja

Cu
pe

o
sa

Sã

de
Ca
have been shown to result more from the social disparities that

Ca
defined by the five shelter deprivations described in Part Two of a result of environmental exposure to hazards and toxins, as

this Report. Where child mortality rates are high, the propor-
tion of slum households is typically also high. In such coun-
tries, child mortality is highest in slums and rural areas and is
lowest in non-slum urban areas.
they tend to have greater contact with the soil and contami-
nated water than adults, and, by virtue of their low body
weight, they are more quickly and adversely harmed by any
toxins that they ingest.
o
Ri
shape health than from the quality of the national health system. City Non-slum Slum
Income, unemployment, education level, quality of housing, eat-
ing habits, and the work environment have emerged as major Source: UN-HABITAT 2005, Urban Indicator Programme Phase III.
health indicators related to social inequality in the United Note: Computed from Demographic and Health Surveys (DHS) data 1995-2003.

110 111
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

■
Rubbish dump, Bangladesh TOPHAM PICTUREPOINT

Income distribution is a more powerful
determinant of health and mortality than
the overall wealth of nations.
Kingdom.5 Urban social ecology studies in different developed
countries also show that health is strongly related to access to
medical care. One study found a strong correlation between mor-
tality and income disparities in the United States, where access to
Large cities tend to display wider inequalities than smaller cities
or towns, even in countries with low levels of socioeconomic
inequality, such as Morocco. In Morocco’s capital city of Rabat,
the under-five mortality rate is 2.7 times higher in slums than in
non-slum areas. Cities that display high levels of inequality, such
as Rio de Janeiro, Brazil, and Cape Town, South Africa, also show
huge disparities between slum and non-slum areas. The under-
five mortality rate in Rio de Janeiro’s slums is three times higher
than the rate in non-slum areas of the city, while in Cape Town,
children under the age of five living in slums are five times more
likely to die than those living in high-income areas.
■ Immunization is no substitute for healthy living
conditions
Persistently higher rates of child mortality in low-income set-
tlements and slums than non-slum urban areas point to defi-
ciencies in current approaches to curbing mortality, which have
focused primarily on immunization against deadly childhood
diseases. Three-quarters of all children in developing regions
are now immunized against measles, and immunization levels
are high in both rural and urban areas, including slums.
Immunization has continued to be a prevalent approach to
decreasing child mortality around the world and is supported
by the international community and individual governments,
but immunization alone appears insufficient for children who
live in slum conditions.
In countries in which most children receive the measles vac-
cine, measles-related deaths have dropped dramatically or been
eradicated. The remaining child mortality rates reflect instead
deaths related to illnesses such as pneumonia, diarrhoea, malar-
ia, and HIV infections, with malnutrition as an important con-
tributing factor. In countries that have been successful in
immunizing children, policies to reduce child mortality must
now address the significant environmental and social factors
that contribute to the death of children under five.
Some countries, including Niger, Nigeria, India, Pakistan,
Diarrhoea: The silent killer in slums
Infectious agents enter the body through four main pathways:
air; food, water and fingers; skin, soil, and inanimate objects;
insect vectors; and mother-to-child transmission.7 Children living
in slums are likely to come into contact with contaminated air,
food, water and soil, and to be exposed to conditions in which
parasite-carrying insects breed. Two conditions – pneumonia and
diarrhoea – are prevalent among children in slums and are respon-
sible for a large proportion of child deaths, each killing more than
2 million children in developing countries each year.8 Despite
their impact on children’s health, pneumonia and diarrhoea – and
the conditions within children’s living environments that cause
them – are not typically given high priority in interventions
aimed at reducing child mortality. In some Asian countries, slum
dwellers are more likely to suffer from diarrhoea than both the
non-slum and the rural population. For example, in Bangladesh,
the prevalence of diarrhoea among slum dwellers is 25 per cent –
double the rural and the non-slum level.
One factor that may explain the high level of diarrhoea in
slum areas is the existence of pit latrines sometimes shared by
hundreds of families. The use of ventilated pit latrines alone is
not a health hazard – in rural households, they have an insignif-
icant relationship to the prevalence of diarrhoea – but in urban
areas, the number of latrines may not be sufficient for the num-
ber of households, leading to unsanitary conditions that
increase the risk of coming into contact with contaminated fae-
cal matter and spreading the bacteria that cause diarrhoea. In
Nigeria and Cameroon, the use of pit latrines in urban areas is
strongly related to the prevalence of diarrhoea. This opens the
debate on whether the current practice of defining ventilated
pit latrines as an acceptable form of “improved sanitation” in
urban areas is still valid.9
FIGURE 3.3.3 DIARRHOEA PREVALENCE AMONG CHILDREN UNDER FIVE YEARS IN SELECTED COUNTRIES
25
Another contributor to the high rates of diarrhoea in slums
may be the fact that in many cities, slum households are not
connected to municipal drinking water supplies. Families may
have to rely on water sold by vendors or from other sources that
may be contaminated. In slums, the risk of contamination
from unhygienic latrines, lack of solid waste disposal, poor
drainage, and inadequate wastewater treatment is also high and
can contribute to the spread of a variety of water-borne and
water-related diseases.
The higher the incidence of slums in cities, the greater the
prevalence of diarrhoeal infections among the urban popula-
tion. In the Sudanese capital of Khartoum – where the slum
population comprises 80 per cent of the urban population, the
highest slum incidence for a capital city – the prevalence of
diarrhoea is 33 per cent, compared with 29 per cent in rural
areas. In Khartoum’s slums, the prevalence is even higher, at 40
per cent. The importance of disaggregating urban data is illus-
trated by statistics collected in Nairobi, Kenya. In 1998, the
prevalence of diarrhoea among slum children in Nairobi was 27
per cent, compared with 19 per cent in rural areas. Official fig-
ures, however, show that the average prevalence of diarrhoea
among children in Nairobi is 12 per cent, a figure that masks
the high proportion of children suffering from diarrhoea in the
city’s slums.
The incidence of diarrhoea among children living in slums is
higher than that of rural children, regardless of household
income. Demographic and Health Surveys show that children
from the highest-income groups within slums have higher rates
of diarrhoea than children of the poorest rural families. This
suggests that the living environment of slum children, in which
they are exposed to contaminated water, soil and air, is a more
important determinant of whether or not a child will have diar-
rhoea than the ability of his or her parents to afford health care.

medical care is often dependent on the ability to pay. However, in
Australia, Canada and Sweden, where medical care is more afford-
able and is often provided for free to vulnerable groups, mortali-
ty was not related to income.6
Degrees of socio-economic inequality and corresponding child
and Haiti, report that measles is still among the five main caus-
es of child deaths, particularly in slums and rural areas. In these 20
countries, the reduction of high mortality rates will require
substantial resources to immunize children against measles as
15
well as improve living conditions that contribute to the inci-
Percentage (%)

mortality rates vary throughout the world. While Northern
African countries, such as Morocco and Egypt, report low average
child mortality rates in a context of few social inequalities, in
Latin America, Brazil displays low overall child mortality rates in
a context of high degrees of socio-economic inequality. In
Morocco, child mortality is only 24 per cent higher in slums than
in non-slum areas of cities, while in Brazil, child mortality rates in
slums are twice the non-slum rates and are comparable to
dence of diarrhoea, pneumonia and malaria. The immuniza-
tion coverage in Niger is as low as 33 per cent, and the coun- 10
Rural
try’s child mortality rate is among the highest in sub-Saharan
Slum
Africa (270 per 1000 live births), with wide inequalities in cov-
5 Non-slum
erage between non-slum (86 per cent), slum (35 per cent) and
rural areas (28 per cent). In Haiti, coverage is comparable to
some of the poorest countries in sub-Saharan Africa, at around 0
50 per cent for both rural and slum areas. Overcrowding, inad-
a

da

il

ia

es
pa
az
ric

ni

bi

bi

al

gu

di

n
an

ne
za

em

Ne

In
Br

pi
ra
Af

Za

lo
Ug
n

do

ilip
slum/non-slum ratios in countries such as Ethiopia. Brazil has

ca
equate water and sanitation and poor hygiene all contribute to

at
Ta
h

Co
ut

In
Gu

Ni

Ph
So

managed to decrease child mortality with advances in its public the prevalence of infection and disease among children. Each of
health system, but children living in slums are still at much greater these environmental factors is more prevalent in slums than in Source: UN-HABITAT 2005, Urban Indicators Programme, Phase III.
risk than their non-slum counterparts. non-slum urban areas. Based on Demographic and Health Surveys 1995-2003.

112 113
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

Battery factory in Dhakar slum, Bangladesh JORGEN SCHYTTE / STILL PICTURES

■ Indoor air pollution and acute respiratory
infections contribute to child mortality in
slums
The concept that the living environment has a direct relation-
ship to child mortality is further supported by data on air pol-
lution and acute respiratory infections. Acute respiratory infec-
tions, primarily pneumonia, account for about 18 per cent of
deaths among children under five.10 Effective treatment with
oral antibiotics can help in preventing bacterial infections that
cause pneumonia, but many poor families in developing coun-
tries cannot get to health facilities or purchase medication fast
enough. Data from 29 countries shows less than half of the chil-
dren with respiratory infections are taken to health care
providers. In West Africa, less than one-third of infected chil-
dren have access to health care providers and the medication
they need to survive.
In cities of both the developed and the developing world, high
levels of indoor and outdoor air pollution caused by motor vehi-
cles, industrial emissions and use of solid fuels for cooking have
led to an increase in respiratory illnesses. According to some
reports, more than 400,000 people die each year in the Chinese
capital of Beijing from pollution-related diseases.11
In slums, high exposure to indoor air pollution caused by use
of solid or biomass fuels, poor ventilation and overcrowding
have also led to higher rates of respiratory illnesses. Families liv-
ing in overcrowded, poorly ventilated housing without adequate
sanitation and safe water are constantly exposed to infectious
air-borne diseases.12 The prevalence of acute respiratory illness-
es is, therefore, much higher in slums and rural areas than in
non-slum urban areas.
Exposure to indoor air pollution from the combustion of solid
fuels, which especially affects women and small children who Endnotes
are more likely to spend more hours indoors, has been implicat-
ed, with varying degrees of evidence, as a causal agent of several 1 United Nations 2005b.
diseases in developing countries. Every year, 1.6 million people 2 Cornia & Mechinii 2005.
die from exposure to indoor air pollution, 1 million of whom 3 Wilkinson & Marmot 2003.
are children.13 4 World Bank 2005.
Approximately one half of the world’s population relies on 5 Crombie, et al. 2005.
biomass – wood, charcoal, crop residues, and dung – and coal 6 Ross, et al. 2001.
as the primary source of domestic energy for cooking and heat-
7 Henry & Chen 1983.
ing. Solid fuel use is especially common among low-income
8 World Health Organization 2002.
households in Africa and South-Eastern Asia. Slum dwellers are
9 UN-HABITAT is currently working with agencies such as WHO and
up to 10 times more likely to use solid fuels for cooking than
those living in non-slum areas. Indoor air pollution can also lead UNICEF on redefining “improved sanitation” in urban areas on the
to illness in non-slum households that have enclosed, poorly basis that while a pit latrine may be a safe and adequate form of
ventilated cooking areas and are situated among other house- sanitation in a rural setting, in overcrowded slums where sharing of
holds using solid fuels. latrines by dozens of people is common, they are neither safe nor
In overcrowded areas, the potential for viral, bacterial, fun- adequate.
gal, and parasitic epidemics is also high. Diseases such as 10 World Health Organization 2002.
meningitis childhood tuberculosis and adult respiratory infec- 11 Watts 2005.
tions appear to be closely associated with overcrowding in 12 See also Henry & Chen 1983.
Incomplete and inefficient combustion of solid fuels results in deprived areas. Conclusive evidence from a study conducted 13 United Nations Department of Economic and Social Affairs, Statistics
the emission of hundreds of compounds, many of which are in São Paolo, Brazil, demonstrated a strong relationship
Division 2005.
health-damaging pollutants or greenhouse gases that contribute between tuberculosis and household overcrowding, particu-
to global climate change. Linkages among household solid fuel larly in smaller housing units, which suggests that the disease
use, indoor air pollution, deforestation, soil erosion and green- requires prolonged contact. Poor ventilation and crowded liv-
house gas emissions have become increasingly important in ing conditions predispose household members to respiratory
understanding the impacts of domestic energy use on the local and skin infections. The lack of a separate kitchen and the use
and global environment, and on health. In addition to their of solid fuel for cooking in an overcrowded house contributes
local and global environmental impacts, biomass and coal to the high prevalence of acute respiratory infections in slums
smoke contain a large number of known health hazards. and urban areas in general.

FIGURE 3.3.4 PROPORTION OF UNDERWEIGHT CHILDREN UNDER FIGURE 3.3.5 PROPORTION OF CHILDREN UNDER AGE FIVE FIGURE 3.3.6 PREVALENCE OF ACUTE RESPIRATORY ILLNESSES AMONG CHILDREN IN SLUMS, RURAL AREAS AND NON-SLUM URBAN
AGE FIVE IN SELECTED AFRICAN CITIES WITH ACUTE RESPIRATORY INFECTIONS IN SELECTED LATIN AREAS, IN SELECTED COUNTRIES
AMERICAN AND ASIAN CITIES

25 50 35
Percent of malnourished children under-five

30
Percentage of children underweight

20 40

25
Percentage (%)

15 30
20

20 15
10
Rural
10 Slum
5 10 Non-slum
5

0 0 0
iro

ut

ca

rto es

sA o
ba

et

Ka s
r e ala

am

a
re
o
em leza

ty

M a

Ko i

Ka a

m i
Ta ad

Ja t
B rta
ro ung

illa

bu
ba

Is ach

go
v

cr

ol
ul

ak

da

il

ru

sh

s
ci

sy

zr
lko

ke

ra

pa
F
No
an

ba
ab

Ce

Ca

az

ne
ka

ric

ni

bi

al

gu

di
an

Da mp

al
Nd
Pa

um

Ac

La
Dh

Na

Pe
M and

de
rta

As

Ha

an
a

sh

za

em

Ne

In
sS
bl

Br

pi
ra
al

Af
o

Za

la
Ug
sa
Fo

la

ilip
ca
ol

at
di

Ta
h

ng
Po
Ca
Sa

ut

Gu
Ad

Ni

Ph
et
at

Ba
So
Gu

Source: Demographic and Health Surveys 1995-2003.

Source: Demographic and Health Surveys 1995-2003. Note: Acute respiratory infections recorded in the two weeks before the survey. Source: Demographic and Health Surveys 1995-2003.

114 115
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

3.4 HIV/AIDS and Urban Poverty

W
here the HIV/AIDS pandemic is ram-
pant, it most deeply affects three linked
populations: the mobile, the urban and
the poor. Halting and reversing the
spread of the disease will require special
attention to the needs and struggles of those living in poverty
in cities – the most vulnerable, and the most at risk.
Historically, migration has served as a major contributing
factor to the spread of HIV/AIDS. Communicable diseases
usually spread faster and farther as road and transport networks Kibera, Nairobi SEAN SPRAGUE/STILL PICTURES

expand; in fact, disease patterns often follow major highways,

seaports and airports.

Safe Sex Campaign, Accra, Ghana DUCLOS-UNEP/STILL PICTURES

In the case of HIV/AIDS, trends indicate that the disease
first appears in cities and then diffuses to rural areas along
major road networks. In Côte d’Ivoire, for instance, HIV first
appeared in the capital city, Abidjan, then spread outwards to
villages throughout the country. In Southern Africa,
HIV/AIDS has been known to spread through transport routes
that begin in Zambia and end in South Africa via Zimbabwe,
Malawi and Mozambique. The highway that stretches from the
Kenyan coastal city of Mombasa through Uganda and Rwanda
has also been cited as an HIV danger zone with blame attrib-
Urbanization has emerged as an
uted mainly to the risky sexual behaviour of truck drivers along increasingly important factor in the
the route. According to a 1994 study, about 50 per cent of spread of HIV/AIDS, particularly in
truck drivers arriving in the Rwandese capital of Kigali from sub-Saharan Africa
Mombasa and Nairobi were HIV positive.1 Tourist centres that
attract travellers and sex workers are also conduits for the trans-
mission of the disease.

■ HIV/AIDS: A largely urban phenomenon
FIGURE 3.4.1 HIV PREVALENCE AMONG MEN AND WOMEN AGED 15-49 IN URBAN AND RURAL AREAS IN SELECTED SUB-SAHARAN
AFRICAN COUNTRIES 2000-2004 In recent years, urbanization has emerged as an increasingly
important factor in the spread of the disease, particularly in
30 sub-Saharan Africa, which is home to more than 60 per cent of
all people living with HIV, or more than 25 million people.2
25 Recent Demographic and Health Surveys in seven African
countries – Burkina Faso, Burundi, Ghana, Kenya, Mali, Niger
20 and Zambia – show that in all countries, HIV prevalence was
Percentage (%)
prevalence nationally, with approximately 2 per cent of the
population infected.3
Data from other regions shows that HIV tends to be concen-
trated in the larger cities, where opportunities for commercial
sex work are higher and where intravenous drug use is more
prevalent. In Argentina, for instance, HIV is concentrated in
the largest cities, and an estimated 65 per cent of HIV infec-
tions occur in the capital, Buenos Aires, alone. Similarly, in
Bolivia, the epidemic is concentrated largely among commer-

higher in urban areas than in rural areas, and was also higher cial sex workers in cities such as Santa Cruz and La Paz.4
15 among urban women than among rural women. In all
Urban countries, women were disproportionately affected, reflecting a
10 Rural general trend in the region. Stark differences were found in ■ Urban poor disproportionately affected
Burundi – the least urbanized country in the region – where
5
the percentage of the urban population infected was almost The urban poor are disproportionately affected by
four times the rural percentage, for both men and women. In HIV/AIDS in both developed and developing regions. HIV
0
Kenya, Tanzania and Zambia, HIV prevalence among urban prevalence appears to be closely linked to levels of poverty in
so

o
ga

nd

da
ge
an

oo

ny

ni

bi

th

populations was almost twice that of rural populations, whereas both the United States and Canada; in the former, the epidem-
Fa

ne

ru

za

an

so
Ni

Ke
Gh

er

Za
Bu
a

n
Se

Ug

Le
m
in

Ta
Ca
rk

in Ghana there was not much difference between rural and ic is disproportionately lodged among African Americans and is
Bu

Source: MEASURE DHS, various surveys 2000-2004. urban prevalence; Ghana also had relatively low HIV affecting increasing numbers of women, while in the latter,

116 117
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
In Latin America, HIV appears to be
concentrated in large cities.
research shows that indigenous people are twice as likely to be
infected as non-indigenous people.5
Heterosexual intercourse accounts for most of the HIV diag-
noses among African American women in the United States, but
the risk factor involves the often-undisclosed risk behaviour of
their male partners. Recent research in a low-income area of New
York City, for example, has shown that women are more than
twice as likely to be infected by a husband or a steady boyfriend
as by casual sex partners.6 This trend was also found in the
Chinese city of Guangzhou, where some 72 per cent of women
with sexually transmitted diseases said they had only had sex with
their husband or regular partner. Women now account for more
than one quarter of new HIV infections in India; most of those
tested said they were in long-term relationships.7
In some countries of Eastern Europe and Central Asia, eco-
nomic transition, liberalization and rising inequality have con-
tributed to the spread of the epidemic. An estimated 1.4 mil-
lion people are living with HIV in these regions – a greater than
nine-fold increase in less than 10 years. The Russian Federation
has the largest number of people living with HIV in the region
and accounts for some 70 per cent of all HIV diagnoses offi-
cially registered in Eastern Europe and Central Asia.8
Intravenous drug use appears to be a more significant factor in
the spread of the disease than commercial sex, although the
prevalence of drug use among commercial sex workers is high,
increasing their chances of contracting the virus.
High levels of unemployment and low wages are related to
transactional sexual activity among the urban poor, and many
women are resigned to using sex to meet their basic needs for
food, shelter and clothing. Low socio-economic status increas-
es the risk of transactional sex among women and raises their
risk of experiencing coerced sex.9 Indeed, economic hardship
not only compounds women’s sexual vulnerability, but is also
associated with early sexual debut and pregnancy, extramarital
sexual activity, and multiple sexual partnerships, all of which
have serious implications for the spread of the disease.10
Clearly, the inability of many people living in poverty in
cities to satisfy their basic needs has implications for the spread
and management of HIV/AIDS transmission, which in sub-
Saharan Africa is primarily via heterosexual relations. High-
risk sexual behaviour, fractured family networks and poor
access to health services appear to account for the high preva-
lence of HIV/AIDS in Africa’s urban areas. The African
Population and Health Research Center’s work with collabora-
tors has shown that the poorest women in Kenya’s capital city,
118
Streetwalker © ALEKSEY KLEMENTIEV. IMAGE FROM BIGSTOCKPHOTO.COM
Nairobi, initiate sex one year earlier than their rural counter-
parts, and three to four years earlier than their wealthier city
counterparts.11
Similarly, the proportion of the urban poor with multiple sex-
ual partners is significantly higher than for the rural poor. This
may be attributable to the extreme poverty in slum communi-
ties interacting with the centrality of money to urban survival.
Because men in urban slums generally have low incomes, they
pay very little for their sexual transactions, in a sense forcing
women to retain multiple partners in order to make ends meet.
More troubling is the fact that, in the face of these realities,
the urban poor often do not use condoms for protection
against sexually transmitted diseases. In India, for instance,
research reveals that one quarter of street-based sex workers do
not use a condom if their clients decline to use one. Despite
their riskier sexual practices, and greater knowledge of and
access to condoms, urban dwellers in Kenya are only slightly
more likely to use condoms than those who live in rural areas
– 10.8 per cent versus 8.3 per cent, respectively.12 These find-
ings are robust even for married women. Nairobi’s poor are
actually less likely to use condoms than the rural poor in
Kenya, further increasing their risk of contracting HIV.
The risky sexual outcomes observed among the urban poor
are not simply a result of low HIV/AIDS awareness. Rather, the
manifestation of deprivation in urban settings appears to disad-
vantage residents more than in rural contexts. High unemploy-
ment, low and unstable wages, small and congested living
spaces, and fractured family and social relationships all con-
tribute to the urban poor’s vulnerability, which forces them to
resort to sexual behaviour they might otherwise avoid.13
Conversely, the rural poor often do not face the same magni-
tude of challenge and survival difficulty as their urban counter-
parts, in the sense that rural residents may have fewer housing
expenses and may be able to grow part or all of their own food.
The deprived conditions in urban slums also serve to
encourage children and adolescents to experiment with sexual
activity at an early age.14 Parents in poor urban settings worry
about their children being socialized into sex at a young age.
Children are exposed to prostitution in their communities and
Slum Conditions Increase Risk of HIV Infection in Nairobi
Given their poor access to proper med-
ical facilities, people living in rural areas
in developing countries are often
assumed to have worse health out-
comes than people living in urban
areas. While this assumption generally
holds true in most countries, in Kenya,
evidence suggests that those in urban
slums are worse off than their rural
counterparts and are more vulnerable
to infectious diseases such as HIV/AIDS.
A survey conducted in 2000 by the
African Population and Health Research
Center (APHRC) found that compared
to other areas in the country, slum resi-
dents in the capital city Nairobi suffer
worse health and reproductive health
conditions than their non-slum coun-
terparts. Not only are morbidity risks
for all major childhood diseases (fever,
cough, diarrhoea) higher for slum chil-
dren compared with children else-
where, but slum children also have less
access to immunization, and subse-
quently suffer higher mortality rates
than children in rural areas.
Vulnerability to HIV/AIDS was also sig-
nificantly high in slums, particularly
among girls and women.
The survey also revealed a marked dif-
ference in perceptions of how to avoid
contracting HIV. Despite being the
most widely known sexually transmit-
ted infection in slum communities,
HIV/AIDS awareness campaigns
appeared not to have reached some of
the most vulnerable women. The
APHRC survey showed that a substan-
tial percentage of uneducated and
never-married women who reside in
the city’s slums (the group most likely
to supplement household income with
commercial sex) were ill-informed
about the disease. The lack of aware-
ness about HIV/AIDS prevention
among the urban poor is reflected in
national data on HIV prevalence.
According to UNAIDS/WHO, in 2002,
Sources: African Population and Health Research Center (APHRC) 2002; Government of Kenya/UNCHS 2001; Central Bureau of Statistics 2000; Nalo 2002; UNAIDS/WHO 2004a;
UN-HABITAT/DPU 2003, Chamber of Justice and others 2005, Daily Nation 2005.
HIV prevalence in Kenya was more than
two times higher in urban areas (14.3
per cent) than in rural areas (6.3 per
cent). As in many African countries, HIV
prevalence in Kenya is also higher
among women (8.7 per cent) than
among men (4.5 per cent). However,
more recent data shows a marked
decrease in infections nationally, from
10 per cent in 2003 to 6.1 per cent in
2005, which is mainly attributed to
increased awareness and the establish-
ment of voluntary testing centres
(VCTs) countrywide.
Poverty-driven commercial sex, crime,
domestic violence, child abuse,
unwanted pregnancies and unsafe
abortions are some of the most socially
damaging consequences of urban
poverty. In Nairobi, urban poverty is
spatially manifested in dehumanizing,
overcrowded slums that lack the most
basic services, including toilets and
health care facilities. Social exclusion
and the breakdown of social support
structures, such as the family and the
community, have contributed to high
crime levels and to increase in sexually
risky behaviour. A recent report indi-
cates that rape and incest are increas-
ingly becoming urban phenomena in
Kenya and are particularly prevalent in
slums.
With an annual growth rate of between
7 per cent and 4 per cent over the last
two decades, Nairobi remains one of
the fastest-growing cities in Africa. The
population has grown more than ten-
fold since 1960 to over 2.8 million peo-
ple today. The growth of Nairobi
reflects a pattern countrywide: with
nearly 40 per cent of its population
already living in urban areas, the coun-
try will become half urban in less than
a decade. Rapid urbanization and a lack
of increase in service provision, poor
urban management, inefficient revenue
collection, and policies that favour the
rich at the expense of the poor have all
contributed to increasing poverty and
slum growth in the city. Official data
shows that while absolute poverty
increased from 48 per cent to 53 per
cent in rural areas between 1992 and
1997, poverty in Kenya’s urban areas
increased by a much larger margin:
from approximately 29 per cent in 1994
to 50 per cent in 1997. An official slum
survey shows that 35 per cent of
Nairobi’s households live in slums,
although other sources estimate a
much higher slum population of
between 40 and 60 per cent of the
city’s total population.
Nairobi’s slums have been described as
among the most dense, unsanitary and
insecure slums in the world. Recent
studies indicate that only 24 per cent of
slum households in the city have
access to piped water; slum residents
pay significantly higher charges for
water than other Nairobi residents,
adding to their financial burden. In
some slums, more than 200 people
share a single toilet.
Social isolation, poor or non-existent
basic services, overcrowding, low
incomes, illegal status, and a generally
dehumanized existence all combine to
make slums in Nairobi vulnerable to a
host of health and environmental haz-
ards, which are manifested in higher
prevalence of HIV/AIDS and other
infectious diseases, including HIV-relat-
ed tuberculosis. Combating HIV/AIDS
and other infectious diseases in slums –
by providing better access to water,
sanitation, adequate housing, health
facilities, information, and education –
will not only improve the health and
dignity of the urban poor, but will also
help them become more economically
productive and improve their liveli-
hoods, leading to a general reduction
in urban poverty in the city.
119
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Extreme Deprivation in Poor Urban Settings in Sub-Saharan Africa
• Because of extreme levels of poverty and the unique social
characteristics of urban poor settings, the urban poor are,
to a large extent, more likely than their rural counterparts
to initiate sex very early and to have multiple sexual
partners.
• Urban poor women initiate sex one year earlier than the
poorest women in rural areas and three to four years
earlier than their wealthiest counterparts in urban areas.
• The proportion of Nairobi’s poorest who engage in multiple
sexual partnerships is more than three times greater that
of the city’s wealthiest residents.
• Married women living in Nairobi’s informal settlements are
at least three times as likely as their rural counterparts to
have multiple sexual partners.
• Even though the urban poor exhibit riskier sexual behaviour,
their condom use rates are low (10.5%), and do not vary
significantly relative to those of the rural poor (8.3%).
Source: African Population and Health Research Center (APHRC).
Big City Life ©GALINA BARSKAYA. IMAGE FROM BIGSTOCKPHOTO.COM
120
Disadvantages Associated with Slum Settings Compared to Rural
Settings
• No privacy exists for parents to have sexual intercourse. This does not only
foster an interest in sexual activity among children at young ages, but also
denies parents moral authority over their children as it relates to sex.
• The financial ability to meet immediate basic needs of food and shelter
overshadows the risk of contracting HIV/AIDS and other sexually
transmitted diseases (i.e., women opt to engage in commercial sex to buy
food for their children despite knowing the dangers to which they are
exposing themselves).
• Because of widespread prostitution in urban poor communities, many
young girls living in these communities consider prostitution a viable
livelihood regardless of its risks.
• There are higher proportions of single men and women in slum
settlements than in any other community, and this contributes
significantly to the high levels of risky sexual behaviours among
residents of these communities.
their parents’ own sexual activities at home, owing to cramped
living quarters and lack of privacy. Parents argue that being
seen or heard having sex impacts their dignity and robs them
of the moral authority over their children. Perhaps most dis-
turbing, the economic deprivation in poor urban communities
appears to have commercialized sex even for adolescent girls,
who have little else to trade but their bodies. When the eco-
nomic situation gets especially desperate, parents sometimes
draft their young daughters into contributing their share of
household expenses.
There are multiple links between poverty and risky sexual
behaviour among young people. Research in Southern Africa
has shown that poorer young people have less knowledge of
HIV/AIDS and begin having sex at younger ages than their
wealthier peers. Poverty and lack of parental resources are
cited as primary reasons for young women to trade sex for
goods or favours or to engage in relationships that involve
financial support. Condom use is reported to be consistently
lower in these types of sexual encounters.15 In a national sur-
vey in South Africa, young people aged 15 to 24 living in
poor informal settlements had more than double the HIV
prevalence of those residing in wealthier urban areas: 20 per
cent versus 9 per cent, respectively.16 In this age group, 79 per
cent living in informal urban settlements reported being sex-
ually active as compared to 53 per cent of those living in for-
mal urban areas. In another large survey in South Africa,
researchers showed that young people in poor informal urban
areas had a much higher HIV prevalence rate than those liv-
ing in urban formal areas: 17 per cent versus 10 per cent,
respectively. HIV prevalence was three times higher among
young women than among young men.17
Endnotes
1 Bwayo et al. 1994.
2 UNAIDS 2004.
The worst orphan crisis is in Africa, where 12
3 Demographic and Health Surveys in these countries conducted
million children have lost one or both parents between 2001and 2003. Data derived from UNAIDS/WHO AIDS
to AIDS; by 2010, this number is expected to Epidemic Update, December 2004.
climb to more than 18 million. Many of these 4 UNAIDS/WHO 2004b.
children end up on city streets. 5 Ibid.
6 McMahon et al. 2004.
7 UNAIDS/WHO 2004b.
8 Ibid.
9 Hallman 2004.
10 Carael & Allen 1995.
■ The socio-economic impact of HIV/AIDS 11 Zulu et al. 2002.
12 Ibid.
The loss of income-earning family members to AIDS has 13 A detailed report of this research is forthcoming in Social Sciences
significant socio-economic implications for the urban poor. and Medicine – “Urban-rural differences in the socio-economic
In slums, where there are large numbers of female-headed deprivation-sexual behaviour link in Kenya” by Dodoo, Zulu & Ezeh.
households, the loss of a parent can be devastating to chil- 14 Dodoo et al. 2003.
dren, who may be forced to drop out of school, become 15 Eaton et al. 2003.
street children or engage in prostitution to meet the needs of 16 Shisana 2002.
younger siblings. 17 Pettifor et al. 2004.
The worst orphan crisis is in Africa, where 12 million chil- 18 UNAIDS 2004.
dren have lost one or both parents to AIDS; by 2010, this 19 Ibid.
number is expected to climb to more than 18 million. Many 20 Ibid.
of these children end up on city streets, where their chances
of escaping poverty are even lower. A recent study in
Cambodia found that one in five children in AIDS-affected
families had to start working to support their families. Many
had to leave school or forego necessities such as food, medi-
cine and clothing.18
At the national level, the epidemic’s economic impact on
societies has been devastating. In sub-Saharan Africa, many of
the worst-affected countries are also among the poorest.
Zambia’s gross domestic product shrank more than 20 per
cent from 1980 to 1999,19 around the same period when
almost a quarter of its urban population and one-tenth of its
rural population became infected with HIV.
The epidemic’s demographic impact is profound: if current
infection rates continue, up to 60 per cent of Africa’s 15 year-
olds will not reach their 60th birthday.20 AIDS threatens eco-
nomic security and development because the disease primari-
ly affects people in the prime of life, between the ages of 15
and 49. The International Labour Organization (ILO) proj-
ects that the labour force in 34 African countries will shrink
by 5 per cent to 35 per cent by 2020 because of AIDS. This
has serious repercussions for the continent’s ability to achieve
sustainable economic growth and poverty reduction. All of
the factors involved in urban poverty must be confronted in
order for cities, countries and the international community to
make progress toward meeting the Millennium Development
Goal 6 target of halting and reversing the spread of
HIV/AIDS.
121
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
3.5 Education and Youth Employment:
Debunking Some Myths about the ‘Urban Advantage’
“If we are serious about reaching the Millennium Development
Goals by 2015, we must involve young people today. We must
invest in them; we must learn from them; we must be their
partners.” - UN Secretary-General, Kofi Annan1
W
e live in a youthful world. Almost half of
the global population is under the age of
24; 1.2 billion people on the planet are
younger than 15.2 While the overall
share of children and youth in the glob-
al population is shrinking as fertility rates decline, in absolute
numbers, there are more young people today than ever before.
Fully 85 per cent of the world’s working-age youth,3 those
between the ages of 15 and 24, live in the developing world –
primarily in Southern Asia and Africa. Within developing
regions, it is the least developed countries that remain
younger than the rest of the world: in 2005, the global medi-
an age was 28 years, but in 10 least developed African coun-
tries, the median age was 16 or younger.4
Youth embody a significant proportion of the world’s
human capital, but more than 500 million of them live on
less than $2 per day. And while more young people are
attending school today than ever before, 113 million chil-
dren are still not enrolled and 130 million youth remain illit-
erate.5
Issues affecting children and youth are often framed as
problems germane to underdeveloped rural areas rather than
cities. Indeed, in general, cities appear to foster the healthy
development of children and youth, providing easier access
to education, health care and employment for young men
and women than is available in rural villages. However, not
all who grow up in cities benefit from the so-called “urban
advantage”, as data collected by UN-HABITAT and its part-
ner agencies reveals. This chapter presents data on the stark
differences for young people within cities: those living in
slums, and those living in non-slum urban areas.6
■ Intra-city inequalities in access to education
Available data indicates that school enrolment rates are in
general much higher in cities than in villages. In countries
such as Burkina Faso, rural communities lag far behind their
122
First Day in School, Patuakhali City, Bangladesh JORGEN SCHYTTE/STILL PICTURES
urban counterparts, with 21 per cent enrolment in rural areas
and 73 per cent in cities. In Burkina Faso, living in an urban
area has a clear advantage, regardless of whether one is rich or
poor. Inequalities in access to school facilities can partly
explain this urban-rural differential, but surveys in other coun-
tries show that while school enrolment rates in rural areas are
dependent on the availability and accessibility of school facili-
ties, the availability of schools in urban areas is not sufficient
cause for children to be enrolled in school. Families in slum
communities, in particular, often cannot afford to send their
children to school because the combined costs of school fees,
textbooks and uniforms are prohibitive. In Kenya, for exam-
ple, the government mandated free primary education in
2003, but students must still purchase uniforms and supplies,
and pay fees to take exams, making it difficult for low-income
families to send their children to school and ensure their
progress. Even in slum areas served by several schools, the
number may not be sufficient, further prohibiting children’s
access to quality education. A study in the Nairobi slum of
Kibera in 2003 found that while 14 public primary schools
were situated within walking distance of the slum, the schools by UN-HABITAT’s analyses of urban survey data. Similar sit- movement to proscribed areas. More barriers to education
could only accommodate 20,000 of the more than 100,000 uations have evolved in Zambia and Zimbabwe, as well, but the exist for girls than for boys around the world. Where resources
primary school-age children living in the area.7 disparity is not confined to sub-Saharan Africa. In Guatemala are limited and school systems are less responsive to the needs
Lack of access to school for poor children in cities is exacer- in 1999, only 54 per cent of children living in slums were of girls, they risk losing important opportunities to fulfill their
bated by the fact that most national and international literacy enrolled in primary education, versus 73 per cent in non-slum potential and improve their lives.
and education programmes have focused in recent years on urban areas and 61 per cent in rural areas. The same situation
reducing the urban-rural gap in education. Although much was observed in Brazil in the late 1990s. Studies indicate that a
remains to be done in rural areas, it is important to recognize majority of parents settling in slums postpone sending their ■ The gender gap in urban education
that in the past decade there has been a significant increase in children, especially girls, to school, until they can manage other
enrolment in rural areas and a decrease in enrolment in impov- expenses, such as food, rent and transport. Eliminating gender disparities in access to education is essen-
erished urban communities. Causes of social inequality in basic education vary from tial to the achievement of the Millennium Development Goals
The problem is evident in poverty-stricken areas of many country to country, but there is a common set of constraints – particularly Goal 3 on promoting gender equality and
African cities, where primary school enrolment is decreasing. In to be considered, including poverty; the embedded costs of empowering women. Girls have historically had less access to
Eastern and Southern Africa, the most significant progress in education; shortage of school facilities; unsafe school environ- educational opportunities than boys in many countries; in
school enrolment in the late 1990s was concentrated in rural ments, especially in poor urban neighbourhoods; and cultural 2005, the United Nations Children’s Fund (UNICEF) found
areas, leaving many poor urban families behind. In Tanzania and social practices that discriminate against girls, including that girls in 54 countries still did not have equal access to basic
net enrolment ratios increased in both rural and non-slum requirements that they provide domestic labour, marry and education.8 Countries in several regions have made progress
urban areas, but actually decreased in slum areas, as indicated have families at a young age, and limit their independent toward the goal of gender parity, however. In the 1990s, the
123
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

Immigrants in Paris: Dreams Go Up in Flames

FIGURE 3.5.1 NET ENROLMENT RATE (PRIMARY) BY TYPE OF RESIDENCE IN SELECTED COUNTRIES
Between April and September 2005,
three fires ravaged residential buildings
100 in Paris, killing 48 African immigrants,
primarily from Senegal, Côte d’Ivoire
and Mali. Most of the victims were chil-
80 dren; many were undocumented.
The immigrants lived in cheap hotels
60
Percentage (%)
many contend that the law is not ade-
quately enforced.
Finding appropriate housing remains
challenging even for families who can
afford market rental rates. In 2002, the
housing vacancy rate in Paris was 6.2
per cent, the lowest since the late 1960s.
are targeted by police and excluded
from opportunity in France, has forced
many immigrants to ask themselves
whether they are really better off in their
adopted lands.
Lack of opportunity and social exclusion
remain major political and social issues

and apartment houses ill-equipped for
emergencies, lacking smoke detectors,
40 fire extinguishers, emergency exits, and,
in one case, even running water with
which to put out the blaze. Some of the
20
families had been placed in the sub-
standard accommodations by social
0
service agencies while waiting for their
residency papers to be processed.
t
Those few units that are vacant tend to
be substantially older than occupied
ones. In the ageing and dilapidated
buildings in which the fires occurred,
only one exit was available – via the cen-
tral wooden staircases, which burned
quickly and left families stranded on the
upper floors.
for immigrant communities in France.
The youth who spoke out during and
after the riots protested vehemently
over two questions employers consis-
tently asked during job interviews: the
applicants’ ethnic origins and their
address. Employers were known to dis-
criminate against those who lived in
stigmatized suburbs. (A similar study in

ria

da

ia

sh

an

am

il

iti
yp

ne

az
cc

ni

oo

ric

di

ta

bi

al
an

Ha
de
an

st
ge
Eg

Be

In

em
kis

Br
pi
o

er

Af

Others entered the tenements on their The Paris city government plans to reno- Rio de Janeiro found that living in a
nz

kh
la

lo
or

Ug
Ni

et
ilip
m

be

at
Ta
h

ng

za

Co
M

Vi
Ca

ut

Gu
Ph
Uz
Ka
Ba
So

own, squatting in the only shelter they vate 1,000 identified substandard apart- favela appeared to be a bigger barrier to
could find. ment blocks, in addition to building gaining employment than being dark
Urban Rural Non-slum Slum
60,000 units of housing each year to skinned or female.) Unemployment
For refugees from African slums seeking help quell the crisis.Tenants’ advocates, among immigrant communities in
Source: UN-HABITAT 2006, Urban Indicator Programme Phase III. a better life in Europe, Paris offers little however, maintain that more than France is estimated to be around 40 per
Note: Computed from Demographic and Health Surveys DHS data 1995-2003.
relief from the insecurity and destitution 120,000 new units are needed each year. cent, 30 per cent higher than the nation-
they experienced at home. Officials esti- For immigrants awaiting both housing al average. A recent study found that
mate that more than 200,000 people are and legal resident status in the tene- white male applicants were 5 times
gender gap in primary school enrolment narrowed, most evi- ing school fees, and daughters are typically the first casualties of homeless or living in temporary shelter ments of Paris, every day in a building more likely to get job offers than those
dently in regions where the gap was wide, such as Northern this choice. Girls are more likely than boys to suffer from lim- in the city. Subsidized social housing with faulty wiring, inadequate plumbing with Arab-sounding names or those
Africa. In the developed regions and in Eastern Asia, the gen- ited access to education, especially in urban poor and rural units are scarce – in 2004, more than and only one way out brings the risk of whose physical home address was
der disparity has reversed, with more girls than boys now areas. Secondary analysis of survey data shows that on average, 100,000 families were on waiting lists for another tragedy. among area postal codes that were
enrolled at the primary level. the single most common reason young women reported for 12,000 available units. Some families lan- deemed “undesirable”.
Progress indicated by regional estimates has been uneven leaving school was inability to pay the associated fees. In the guish in overcrowded and filthy provi- The problems for immigrants in Paris are
within regions. Where girls are still at a disadvantage, urban areas of Uganda and Zambia, for instance, 74 per cent sional dwellings for 14 years or longer deeper than substandard housing, as The disparities in housing and employ-
resources and school facilities are limited and enrolment is and 51 per cent, respectively, of young women between the while they wait to be accommodated in demonstrated by the riots that swept ment opportunities between immigrants
altogether low. In many countries with low overall enrolment, ages of 15 and 24 gave inability to pay as the main reason they social housing. Such long waits are not the city in October and November 2005. (most of whom are French nationals) and
fewer than 50 per cent of primary school-aged girls are stopped going to school. uncommon for immigrants. A govern- Young residents of minority communi- the local population has prompted the
enrolled. Female illiteracy rates are still high in these parts of While primary school tuition fees have now been abolished in ment study found that nearly 30 per ties throughout Paris and its suburbs French government to create more
the world, particularly in urban poor and rural areas, where many countries, public secondary education remains competitive cent of immigrant applications had responded with violence to the acciden- health, education and employment pro-
many girls drop out of school too early to be able to acquire and tuition-based in many parts of the developing world, limit- been pending for more than three years, tal deaths of two teenage boys of grammes aimed at young people living
the necessary skills to function as literate individuals. ing the number of students who can continue their education. two times the national average. African origin, setting cars and buildings marginalized, low-income neighbour-
Demographic and Health Survey data points to four main Even for primary school, nearly all developing countries still ablaze for more than two weeks.The hoods. Stigmatization and exclusion of
reasons why girls discontinue their education: lack of require families to pay fees of various kinds – in many cases, these Although droit au logement, or the right frustration and anger expressed in the neighbourhoods from the rest of society
finances, early marriage and pregnancy, domestic work fees amount to more than the former tuition costs. Fees for uni- to housing, is ensconced in French law, riots grew out of the marginalization of appears to have exacerbated the crisis in
responsibilities, and poor performance. Only a small propor- forms, materials and other educational expenses have been access to a decent, affordable place to ethnic and religious minorities, the Paris. The French city of Marseilles for
tion of girls and young women who had left school – fewer shown to affect girls’ chances of going to school more than boys’, live remains elusive for the lowest- majority of whose members live in run- instance, was immune from the riots
than 10 per cent – indicated that they stopped attending as they add to the already high costs of sending girls to school. income and minority residents. down high rise housing estates in poor largely because the poor are not physi-
because they had graduated. Among some impoverished urban communities, it is common Legislation passed in 1991 requires that neighbourhoods. Growing resentment cally isolated within the city; there low-
for families to choose to educate their boy children in their vil- major cities dedicate 20 per cent of their over unemployment in their communi- income and higher-income communities
Lack of finances lage of origin where schools are less expensive; girls, on the other housing stock to the social sector, but ties and the overriding sense that they are more integrated.
hand, remain in the city to help parents with housework. This is
The direct financial costs of sending all children to school are reflected in the age pyramid of slum areas, which shows that
often too high for families living in poverty in cities. Faced with slum communities have more girls than boys between the ages of Sources: Ford 2005; BBC News 2005b; Bennhold 2005; Norris and Shiels 2004; Langley 2002; BBC News 2005c, TIME 2005; Perlman 2005.
household expenses, urban families may cut back by not pay- 5 and 14 years. (See age pyramids in chapter 1.2, for example.)

124 125
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
FIGURE 3.5.2 PROPORTION OF WOMEN AGED 15-24 WHO STOPPED
GOING TO SCHOOL BECAUSE OF INABILITY TO PAY SCHOOL FEES
Cameroon
Côte d Ivoire
Kenya
Mozambique
Nigeria
South Africa
Uganda
Zambia
Zimbabwe
Colombia
Peru
Indonesia
Philippines
0 10 20 30 40 50
Slum
Source: UN-HABITAT Global Urban Observatory 2005.
Early marriage and pregnancy
A significant proportion of girls who discontinue their edu-
cation in the higher grades of secondary school in urban areas
leave school because of pregnancy. In many countries, espe-
cially in Eastern and Southern Africa, this proportion is par-
ticularly high. In Zambia, 17 per cent of the girls who
dropped out of school in urban areas did so because they were
pregnant; in the Central African Republic, 16 per cent of
female dropouts cited pregnancy as the reason for leaving
school, as did 12 per cent of female dropouts in Uganda’s
urban areas. Sexual harassment and abuse in schools further
impacts the dropout rate; in some places, parents are inclined
to withdraw girls from school to prevent them from getting
pregnant or contracting HIV. In some countries, such as
Chad and Nigeria, girls in urban areas often stop going to
school to get married: 28 per cent of girls who left school in
Chad cited marriage as the reason, as did 18 per cent of
female dropouts in Nigeria.
Within urban areas, stark differences exist. Young women
and men residing in slum areas are more likely to have a child,
be married or head a household than their counterparts living
in non-slum areas in cities. In Uganda, for instance, a 1999
Demographic and Health Survey revealed that 34 per cent of
young people in slum communities headed households, ver-
sus five per cent in non-slum communities. The same pattern
was observed in Kenya (27 per cent versus 14 per cent) and
Côte d’Ivoire (16 per cent versus two per cent). The preva-
126
lence of HIV/AIDS in poor urban communities in these ■ Youth are employed in the growing informal
countries may be contributing to their higher rates of youth- sector Young women and men residing in
headed households.9 slums are more likely to have a child, be
In cities of the developed world, more jobs are being created married or head a household than their
Domestic work in the financial sector and in information management as a
counterparts living in non-slum areas.
result of globalization, while in the developing world, trends
Some young women stop going to school to help their fami- point toward an increasing “informalization” of the urban
lies with domestic chores, including taking care of children. This economy, as the formal sector fails to provide adequate employ-
phenomenon is particularly prevalent in slums where, in the ment opportunities for the number of young people and adults majority of young people working in the urban informal sector
absence of extended family, girls are taken out of school to do seeking work. According to the International Labour live in slum areas. For example, in Benin, slum dwellers com-
domestic work, such as fetching water, while their parents strug- Organization11, approximately 85 per cent of all new employ- prise 75 per cent of informal sector workers, while in Burkina
gle to earn an income for food, housing and other necessities. In ment opportunities around the world are created in the infor- Faso, the Central African Republic, Chad and Ethiopia, they
Mali and Chad, more than 10 per cent of young women in slums mal economy. In some countries, employment in the urban make up 90 per cent of the informal labour force.
cited “helping the family” as the main reason why they stopped informal sector has risen sharply over the past decade.
going to school. Family demands on girls’ time place many Lithuania, for example, experienced a 70 per cent increase in
obstacles in the way of gender equality in access to education. urban informal employment as a percentage of total employ- ■ Gender differences in employment
Other studies show that going to school is seen as a hindrance to ment between 1997 and 2000. The Economic Commission for
the performance of household chores; parents perceive the cost Latin America and the Caribbean estimates that urban infor- In slum communities, early involvement in family responsi-
of lost labour to be greater than the cost of keeping girls out of mal employment in that region increased from 43 per cent in bilities may explain the high employment rates of young men
school. These perceived opportunity costs are usually much high- 1990 to 48.4 per cent in 1999.12 and the low employment of young women. Youth residing in
er for girls than for boys, since girls are expected to do more The informal economy can afford youth a necessary pathway slum areas are more likely to have a child, be married or head
domestic work than boys. By the age of 10, girls in Bangladesh to legitimate work by conferring experience and self-employ- a household than their counterparts living in non-slum areas.
60 70 80
and Nepal may be working up to 10 hours a day in productive ment opportunities. Tracking how many youth participate in In Uganda, 34 per cent of young men living in slum areas head
Non-slum activity inside and outside the home, while Ethiopian girls of pri- the informal sector is difficult for a number of reasons, howev- a household compared with 5 per cent of young men living in
mary school age often work 14 to 16 hours a day.10 er, and limited data currently exists.14 But some trends are non-slum areas. Family responsibilities at a young age often
beginning to emerge. UN-HABITAT analyses indicate that the compel young men to seek and obtain jobs.
Poor performance
The combined social and cultural factors that make it difficult
for girls to enroll in and complete school also contribute to their
dissatisfaction with and poor performance in school. Domestic
responsibilities, marriage and motherhood, and financial con-
straints present strong challenges to girls’ ability to maintain reg-
ular attendance and succeed when they do attend. Surveys indi-
cate that a significant proportion of young women in urban
areas stop going to school because of poor performance; the
obstacles they face induce many to drop out before they com-
plete their education or pass key national examinations.
A significant proportion of young women drop out because
Education in Africa UNEP/STILL PICTURES
they “do not like school”. This is the case for more than 30 per
cent of the young women in slum communities in Mali and
Guatemala who had left school. In Egypt, Nicaragua, Central
African Republic, and Burkina Faso, more than 20 per cent of
the young women in slum communities who dropped out of
school reported that they did so because they did not like
school. Schools in many countries are not girl-friendly and in
some cases, they are even hazardous for girls. Failure to provide
adequate sanitary facilities, such as toilets and running water,
causes inconvenience for boys, but can make the situation disas-
trous for girls. Menstruating girls will not attend school if basic A study in the Nairobi slum of Kibera found that while 14 public primary schools were
toilet facilities are not available. Even when toilets are available, situated within walking distance of the slum, the schools could only accommodate 20,000
they are often poorly serviced and maintained, as is the case in
of the more than 100,000 primary school-age children living in the area.
30 per cent of schools in India. In many places, schools fail to
provide separate toilet facilities for boys and girls, putting girls
at risk of sexual harassment.
127
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

populations of young men in urban areas may feed political

FIGURE 3.5.3 PERCENTAGE OF YOUNG WOMEN AND MEN WORKING IN THE INFORMAL SECTOR IN SELECTED AFRICAN COUNTRIES FIGURE 3.5.5 PERCENTAGE OF YOUNG WOMEN AND MEN WORKING
and ideological unrest and provoke violence. As demonstrat-
IN THE INFORMAL SECTOR IN SELECTED ASIAN COUNTRIES
ed by the riots in Paris in late 2005, high youth unemploy-
100 80
ment, particularly within marginalized ethnic minorities, can
create urban unrest, which can challenge government author-
70 ity and endanger national stability. More importantly, high
80 levels of unemployment among youth, particularly in urban
60 areas, indicate that cities are unable to absorb labour, which
in the long term has a direct impact on economic growth and
60 50
poverty reduction.
Percentage (%)

Percentage (%)
40
40
30

20 20

10
0
0

sh

sh

an
s

y
y
ta

al
l

ne

ne
Fa ina

Fa ina
a

a
Iv ôte

an ic

an ic
Iv ôte
da

da
da

da

pa

e
a

a
n

ke
ria

ria

de

de

st
an

an

p
s

rk
bi

bi
bo

bo
nz bl

nz bl

pi

pi
kh

Ne
an

an

Ne
an

an

kh
so

so
rk

rk
ia

ia
re

re
ge

ge

la

la
C

C
m

Tu
Tu
Gh

Gh
Ta pu

Ta pu

ilip

ilip
Ga

Ga

ng

za

ng
Endnotes
Bu

Bu

za
oi

oi
Ug

Ug
Rw

Rw
Za

Za
Ni

Ni
of Re

of Re

Ph

Ph
Ka

Ka
Ba

Ba
d'

d'
d

d
ite

ite
Un

Un
Men Women
Rural Slum Non-slum 1 United Nations Secretary-General Kofi Annan, in his introduction to
Men Women
the exhibition, “Chasing the Dream: Youth faces of the Millennium
Rural Slum Non-slum
Development Goals”. New York, 12 August 2005.
Source: UN-HABITAT 2005, Global Urban Observatory. 2 United Nations General Assembly Economic and Social Council 2004.
Source: UN-HABITAT Global Urban Observatory 2005. 3 “Youth” are defined by the United Nations as all persons between the
ages of 15 and 24. In this Report, “young people” refers to all persons
On the other hand, young women living in slums are less
likely to seek paid employment, as early marriage and child- younger than 24, “children” refers to persons between birth and 15,
FIGURE 3.5.4 PERCENTAGE OF YOUNG WOMEN AND MEN WHO HAVE FAMILY RESPONSIBILITIES IN SELECTED COUNTRIES bearing forces them to stay at home. Six out 10 young women and “youth” follows the conventional definition.
living in Uganda’s slum communities have a child or are mar- 4 United Nations Department of Economic and Social Affairs,
80 ried – double the number in non-slum communities. The Population Division, 2005.
majority of young women in slums tend to have children at an 5 Ibid.
70 earlier age than their non-slum counterparts. In the absence of 6 Inequalities in education and employment are assessed in this Report
an extended family to help with taking care of children, the with data from Multiple Indicator Cluster Surveys administered by the
60 sick and the elderly, young women living in slums are more United Nations Children’s Fund (UNICEF) and Demographic and Health
Percentage of youngwomen and men

likely to stay at home to look after children and do household

who have family responsibilities

50
Surveys, funded by the U.S. Agency for International Development
chores. This limits their opportunity to look for jobs away
(USAID), collected between 1995 and 2003. These surveys include
from home, particularly in the formal sector.
40 items on school attendance and literacy, youth employment and child
labour that can be disaggregated by type of residence: urban and
30
■ Consequences of youth unemployment rural; slum and non-slum. UN-HABITAT has also used some data from
Labour Force Surveys, funded by the International Labour
20
When youth seeking work fail to find productive, decent Organization, and Living Standard Measurement Surveys, funded by

10
livelihoods, they can become socially excluded and enter a cycle the World Bank.
of poverty, experiencing high rates of unemployment across 7 Lugano & Sayer 2003.
0 their life spans.13 The importance of helping youth find pro- 8 UNICEF 2005.
ductive and decent employment has therefore become a pri- 9 See chapter 3.4 for an analysis of the impact of HIV/AIDS on slums.
a

ia

da

ia

an

ia

ia

da

ia

an

s
ne

ne
pi

an

ny

bi

bw

gu

an

ny

bi

bw

gu
an

liv

op

an

liv
an

st

an

st
o

m
Ke

pi

Ke

pi
Gh

ra

Gh

ra
Bo

Bo
hi

nz

ba

kh

hi

nz

ba

kh

mary motivation of international youth policymaking and

Za

Za
Ug

Ug
ilip

ilip
ca

ca
Et

Et
Ta

Ta
za

za
m

10 Watkins 2000.
Ni

Ph

Ni

Ph
Zi

Zi
Ka

Ka

development efforts. 11 International Labour Organization 2004.

Women Men Many countries in the developing world are experiencing
12 Ibid.
distinctive “youth bulges”, which occur when young people
Slum Non-slum 13 International Labour Organization 2004.
comprise at least 40 per cent of the population. There has
been increasing concern among policymakers that the frustra-
Source: UN-HABITAT 2005, Global Urban Observatory. tions accompanying long-term unemployment among large

128 129
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
3.6 Cities: The Front Lines in the Battle for Sustainability
Protest against traffic pollution, Turin, Italy ANGELO DOTO/UNEP/STILL PICTURES
A
s urbanization continues unabated, the global
community1 is confronting the need to think
creatively about cities and their potential for
leadership in harmonizing human settlements
with ecological preservation and sustainability.
Sustainable cities – those that enable all of their residents to
meet their own needs and prosper without degrading the natu-
ral world or the lives of other people, now or in the future2 –
are products of careful planning in the context of their resi-
dents’ daily lives.
Issues of sustainability are being addressed differently in dif-
ferent parts of the world, according to the policymaking and
environmental priorities of cities and countries. In cities of the
developed world, energy consumption remains a major con-
cern, and many urban areas are being redeveloped with an
emphasis on compact neighbourhoods, clean transportation
options and the use of green technologies. Cities such as
130
Coco Taxis, Havana, Cuba ©ALEX BRAMWELL. IMAGE FROM BIGSTOCKPHOTO.COM
Vancouver are also analysing their “ecological footprints”3 and
adopting strategies to reduce their dependence on outside sup-
pliers, such as increasing urban agriculture and localizing more
of their food supply.4 In cities of the Commonwealth of
Independent States, with economies in transition, the priority
is tackling the legacy of underused urban areas, decaying infra-
structure and deteriorating housing stock. Some countries in
Central Europe are also addressing air pollution and health by
imposing heavy taxes on leaded fuel and phasing out its use
altogether.5
In cities of the developing world, the need to accommodate
rapid urban growth, provide essential infrastructure and servic-
es, control air pollution (especially in the rapidly industrializing
cities of Asia) and improve the living conditions of the urban
poor are emerging as new challenges.6 Some cities, such as
Singapore and Curitiba, have adopted careful urban planning
and “greening” policies to significantly reduce air pollution and
the use of private motorized transport. Singapore has been so
successful at preserving its old-growth tropical rainforest, pro-
tecting and planting green spaces, and promoting clean rapid
transit that it has become the only large city in the world that
acts as a carbon sink, soaking up more carbon dioxide than it
Although cities have been much produces.7 Elsewhere, in places such as Thailand, governments
maligned as generators of waste and are embarking on major slum upgrading programmes that will
pollution, consumers of vast amounts also have a positive impact on the urban environment.
of the world’s natural resources and The most innovative cities in the world fulfill the ideals of
contributors to overall environmental Millennium Development Goal 7 – ensuring environmental
sustainability – by integrating environmental stewardship and
degradation, examples from around urban planning to achieve long-term stability and success.
the world demonstrate that cities have Although cities have been much maligned as generators of
the potential to combine safe and waste and pollution, consumers of vast amounts of the world’s
healthy living conditions with natural resources and contributors to overall environmental
remarkably low levels of energy degradation, examples from around the world demonstrate that
cities have the potential to combine safe and healthy living con-
consumption, resource use and waste. ditions with remarkably low levels of energy consumption,
resource use and waste. Cities also offer enormous environmen-
tal opportunities and advantages.8
Cities provide economies of scale. High densities mean low
per capita costs for the provision of piped water, water treat-
ment and collection and disposal of garbage and human waste.
The high concentration of people in cities can lead to a
reduced demand for land relative to population. Urban areas
take up less than 1 per cent of the total land in most countries
(and approximately 3 per cent of the earth’s total surface area).
Although urban sprawl is encroaching onto agricultural land in
some nations, this can be avoided with coordinated urban and
regional planning and effective land use management.
Cities offer great potential for limiting the use of motor vehi-
cles if combined with adequate development of public trans-
port systems. High concentrations of commuters make energy-
efficient mass transit viable and affordable, and proximity
ensures that more trips can be made on foot or by cycling,
given the appropriate infrastructure.
The concentration of production and consumption in cities
offers a range of possibilities for the efficient use of resources –
through reclamation and wastewater recycling, for instance.
Cities that are unable to integrate economic growth with
good planning and environmental care, on the other hand,
can and do pollute the environment, contribute to the reduc-
tion of biodiversity, undermine the natural resource base, and
increase the scale and depth of poverty. Many cities face chal-
lenges to implementing long-term plans for sustainability.
Some of the most serious challenges centre on economic sus-
tainability and poverty reduction, environmental degrada-
tion, social injustice and exclusion, and failures of gover-
nance.9 These challenges are linked to specific problems with-
in cities that preclude their ability to improve the built and
natural environments for their residents. Urban data on sus-
tainability indicators reveals the scope of the issues world-
wide. A common thread through all of the research is the
importance of engaging the urban poor: cities that do not rec-
ognize the impact of environmental problems on their poor-
est citizens, or the environmental costs of unplanned develop-
ment, remain unsustainable.10
131
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Osaka, Japan BINSYO YOSHIDA/UNEP/STILL PICTURES
■ Air pollution
The highly urbanized developed regions of the world are
responsible for the greatest per capita emissions of greenhouse
gases caused by burning fossil fuels. In 2002, people in the
developed regions emitted 12.58 metric tons of carbon dioxide
(CO2) per capita, compared with 2.07 metric tons per capita
in the developing world.11 Heavy use of motor vehicles is large-
ly to blame; in Canada, for instance, cars and trucks were the
largest source of greenhouse gas emissions in the country in
2002, contributing 25 per cent of such emissions.12 Even as
developed countries work to limit air pollution, global emis-
sions of CO2 – the principal greenhouse gas – are predicted to
rise by more than 60 per cent in the period between 1997 and
2010. The greatest increase – 65 per cent – will come from
developing countries, and primarily from China.13
In the rapidly industrializing cities of Asia, ambient air pol-
lution is on the rise as industrial and motorized transport emis-
sions increase, and as dependence upon fossil fuels persists.
China is home to 16 of the world’s 20 most air-polluted cities
and is the world’s second-largest producer of greenhouse gases
after the United States. Over the past 10 years, the concentra-
tion of pollutants in China’s air has increased by 50 per cent.
Urban outdoor air pollution, mainly from vehicle exhaust and
industrial emissions, is responsible for the deaths of 3 million
people around the world each year – most of them in develop-
ing countries.14 In Beijing alone, more than 400,000 people die
each year of pollution-related illnesses.15 China is the largest
132
producer and consumer of bituminous coal – the main contrib-
utor to its air pollution – and more than 64 per cent of its cit-
izens use coal in their homes.16
Indoor air pollution from the burning of biomass fuels – fire-
wood, charcoal, crop residues and animal dung – is another
major challenge to environmental health and sustainability in
developing countries, and is a growing problem in cities.
Almost half of the world’s population, 2.4 billion people,
depend upon biomass fuels for their daily energy needs, nearly
all of them in developing countries. That number is expected
to rise by 200 million by 2030.17 The burning of biomass
releases toxic gases and compounds into the air, including car-
bon monoxide and methane, leading to a host of chronic respi-
ratory diseases, lung cancer and pneumonia in those exposed to
the smoke and particulate matter.18 Women and children are
disproportionately harmed by the burning of biomass, as they
are more exposed to the dangers of indoor air pollution caused
by cooking using fuels that emit toxic gases and particulates.
Every year, 1.6 million people die from exposure to indoor air
pollution, 1 million of whom are children.
UN-HABITAT analyses have shown that the prevalent use of
biomass or solid fuels in poorly ventilated slum households has
increased acute respiratory illnesses among children in Asia and
Africa, where the use of solid fuels among low-income house-
holds is common. The per capita proportion of biomass use is
highest in sub-Saharan Africa, but the greatest numbers of peo-
ple who depend on the highly polluting fuels live in China and
India.19 In the slums of sub-Saharan Africa and Asia, use of
solid fuels among urban households with extreme shelter dep-
rivations averages 74 per cent and 60 per cent, respectively.
Adopting cleaner-burning charcoal briquettes made of recycled
ash and agricultural waste, along with more modern cooking
equipment, could prevent up to 2.8 million premature deaths
each year.20
■ Traffic deaths
Motor vehicles are not only a major cause of air pollution –
they are also responsible for most fatal accidents in urban areas.
Traffic deaths are symptomatic in many places of failed urban
planning and inadequate roads and transport systems. In the
developing world, more and more motor vehicles are crowding
onto roads not designed to handle them. Where people have
limited transport choices, private cars and shared taxis may be
the only viable options for getting where they need to go, put-
ting them at risk of accidents. Since 1990, there has been a
four-fold increase in the number of motor vehicles in China
and Thailand – a common trend across the developing world.
Increased auto, bus and motorcycle traffic has led to higher
rates of accidents and fatalities in developing countries, where
pedestrians, bicyclists and traditional vehicles share the roads.
Global traffic deaths rose from approximately 990,000 per year
in 1990 to nearly 1.2 million per year in 2002, with 85 to 90
per cent of the fatalities occurring in low- and middle-income
countries.21, 22
In Europe and North America, traffic deaths have been
declining since the 1970s, but they have risen sharply in Latin
America, Asia and Africa. Between 1975 and 1998, traffic fatal-
ities increased by 237.1 per cent in Colombia, 243 per cent in
China and 383.8 per cent in Botswana. By 2020, if the current
trend continues, traffic deaths will increase by 83 per cent in
the world’s low-income countries, even as they decrease by 30
per cent in high-income countries. Many who die in traffic-
related accidents are pedestrians – between 1977 and 1994, 64
per cent of the traffic fatalities in the city of Nairobi were
pedestrians. Better urban infrastructure, pedestrian-friendly
streets and well-planned transport systems that provide safe
options for getting around the city are needed to curb the rise
in traffic deaths.23, 24
■ Inadequate access to water and sanitation
Ninety-five per cent of the world’s urban population had
sustainable access to an improved water source, and 81 per
cent had access to improved sanitation, in 2002.25 The global
numbers are misleading, however – access to safe drinking
water and decent toilets is not evenly distributed among pop-
ulations within regions or even within cities, and gaining
access may involve hardship or risk for residents.26 Households
without adequate water supply and sanitation suffer dispro-
portionately from water-borne or water-related diseases; more-
China is home to 16 of the world’s 20
most air-polluted cities and is the world’s
second-largest producer of greenhouse
gases after the Unites States.
over, lack of these services within slums contributes to the
degradation of water and land resources within cities.
Water and sanitation are intimately linked – where inade-
quate sanitation facilities exist, water contamination is com-
mon. This became startlingly clear when a cholera epidemic
swept East Africa in 1997 and 1998, as a result of human
waste contaminating water sources. The disease started in
slums, where rainwater washed accumulated human waste into
boreholes and other water sources and spread quickly through-
out Kenya, Tanzania and Uganda.27 More common than
cholera is the incidence of diarrhoea, which contributes to
rates of child mortality 10 to 20 times higher in areas lacking
adequate water supply and sanitation than in cities with prop-
er provision of services. The crisis is most acute in the cities of
Africa and Asia. As many as 150 million urban residents in
Africa lack adequate water supplies and an estimated 180 mil-
lion people lack adequate sanitation; three-quarters of the
global population without access to water supply, and more
than half of the population without access to sanitation live in
Asia.28
Access to improved water sources often changes over time.
In East Africa, piped water systems in cities have degraded over
the past 30 years – partly as a result of inadequate maintenance
and urban population growth – leaving more households
without reliable access and decreasing their overall water con-
sumption.29 Less water in households correlates with higher
rates of illness, as it makes washing hands, cleaning cooking
utensils and bathing difficult. Even where water is abundant,
however, inadequate delivery systems and unsanitary condi-
tions can lead to contamination and higher rates of illness.
In addition to degradation of urban water delivery systems,
water scarcity results from over-exploitation of sources, which,
in turn, contributes to environmental crises. Mexico City, for
example, depends upon the Mexico Valley aquifer for 80 per
cent of its water supply, but it has so depleted the aquifer that
the land has shifted and the city is sinking.30 In coastal areas,
where most of the world’s largest cities are located, pollution
of water sources is posing major threats to human and ecosys-
tem health. Lima, Peru, is one coastal city that is contributing
to the global problem: it discharges 18,000 litres of wastewater
per second into the Pacific Ocean.31
133
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

Endnotes
Renewable Energy Sources Increase Urban Sustainability

Water Pollution, China ZHAO WEIMING/UNEP/STILL PICTURES

Fossil fuels made cities what they are today, but then, owing to major investments by governments Other cities, including Portland, Oregon, USA, and 1 The most recent UN-led international discussions on this topic were held at
dependence on them has led to high rates of air pol- and private industries around the world. The Adelaide, Australia, have provided incentives and the World Summit on Sustainable Development in Johannesburg in 2002.
lution and greenhouse gas emissions, poor health Renewable Energy Policy Network for the 21st subsidies for “green building” projects that increase 2 Girardet 2004.
among urban residents and environmental degrada- Century (REN21) reports that at least 48 countries the energy efficiency of homes and commercial 3 The ecological footprint is a measure of how much biologically productive
tion. Life depends upon stable ecosystems, and now have some type of renewable energy promotion buildings and mandate the use of environmentally land and water area an individual, a city, a country or a region requires to
cities, as “eco-technical systems”, have the potential policy, including 14 developing nations. In develop- sound building materials. produce the resources it consumes and the waste it generates.
to work in harmony with the natural world rather ing countries such as China, where electricity and 4 Mougeot 2005b.
than to continue depleting its resources. Indeed, they industry are powered primarily by coal, promising One of the primary strategies for decreasing emis- 5 European Environment Agency 2004.
must in order to survive over the long term. renewable alternatives for meeting energy needs sions, preventing climate change and improving resi- 6 UN-HABITAT/DFID 2002.
Switching to renewable energy sources – such as are emerging in the form of wind power and dent health in cities is upgrading transport systems. 7 Hinrichsen 2002.
wind, solar, modern biomass, geothermal, and small methane gas from decomposing solid waste. Efficient mobility is essential to the economic suc- 8 These arguments are drawn from Satterthwaite 1999.
hydro-electric systems – for the bulk of urban energy Renewable technologies now provide 160 gigawatts cess of cities, and mobility is facilitated by transport 9 UN-HABITAT/DFID 2002.
needs is vital to the sustainability of cities. It is also of electricity generating capacity – about 4 per cent systems that are cost-effective, responsive to 10 This argument has been backed by various studies, including one on urban
an essential aspect of meeting the Millennium of the world total – and global investment in renew- changing demands, environmentally sound, and agriculture by Luc J.A. Mougeot of the International Development Research
Water and sanitation are intimately linked – Centre (IDRC) in Canada.
Development Goals, from improving health and sav- able energy topped US $30 billion in 2004. accessible to all residents. Throughout Brazil, the
ing environmental resources to increasing global practice of mixing regular gasoline with 26 per cent where inadequate sanitation facilities exist, 11 United Nations Department of Economic and Social Affairs, Statistics
partnerships and reducing poverty. As developing Government leadership in creating policies around ethanol – derived from locally grown sugar cane – water contamination is common. Division, 2005.
countries adopt clean, efficient, reliable and renew- renewable energy sources is important to their suc- has led to reduced emissions and increased savings 12 Dooley 2002.
able energy sources and upgrade existing urban sys- cess in the market: renewables are used most in of fossil fuels for cities and citizens. Biofuels are 13 International Energy Agency 1996.
tems, they offset many of the challenges to urbaniza- cities and countries with policy-bound targets for gaining prominence worldwide: trains and buses in 14 Figure from the World Health Organization, reported in Kirby 2004.
tion and make sustainable development possible. reducing carbon dioxide emissions and increasing Sweden are running on methane produced by ■ Unsustainable practices most deeply affect the 15 Watts 2005.
reliance on renewable sources. Freiburg, Germany, degrading animal waste; the entire Halifax, Nova urban poor 16 Karasov 2000.
Much of the current discussion about the need for was one of the first cities to adopt targets for Scotia, Canada, metro bus fleet runs on a mixture of 17 Dooley 2004.
reduced reliance on fossil fuels stems from 1997, reduced greenhouse gas emissions, in the late 20 per cent biofuel and 80 per cent regular diesel; It is clear that when city systems fail to manage basic urban 18 United Nations Department of Economic and Social Affairs, Statistics
when 84 countries signed the Kyoto Protocol at the 1980s. Today, the city of 200,000 is a model of sus- and zero-emissions buses powered by hydrogen fuel Division 2005.
issues – such as regulation of pollution-generating industries
Conference of the Parties associated with the UN tainable urban development, featuring city-financed cells are currently in use throughout London and 19 Dooley 2004.
and provision of clean fuels, development of safe transport sys-
Framework Convention on Climate Change. This solar building projects, subsidies for solar power in nine other European cities. In Helsinki, Finland, a 20 Dooley 2005.
tems, provision of safe and adequate water and sanitation facil-
recognition of the human impact on the world’s cli- new construction, and an overall integrated compact urban design makes mass transit viable. 21 World Health Organization 2004.
ities and collection and proper disposal of solid waste – the
mate and environmental resources spurred innova- approach to urban planning that considers the city’s Elsewhere in Europe and North America, transit-ori- 22 Dahl 2004.
most vulnerable residents suffer the greatest hardship. Per citi-
tion around renewable energy technologies that future energy consumption and sources. Freiburg ented development and “smart growth” – high-den- 23 World Health Organization 2004.
zen, the urban poor in developing areas make vastly smaller 24 Dahl 2004.
curb the production of greenhouse gases and pro- has attracted renewable energy research institutes, sity development that facilitates pedestrian activity
resource demands, make much better use of those resources, 25 Access to improved water sources refers to the percentage of the
vide efficient, environmentally sound power. The use companies, consultancies, solar engineers, and and provides easy access to commuter transit
and produce a much smaller pollution load than do their population who use any of the following types of water supply for drinking:
of renewable energy sources has increased since architectural firms that specialize in solar design. options – are helping to halt urban sprawl.
wealthier neighbours.35 Yet, they endure the greatest environ- household connection, public standpipe, borehole, protected dug well,
Sources: Girardet 2004; Martinot 2005a & 2005b; Balfour 2005; The Economist 2005; Franks 2005; Halifax Regional Municipality 2004; BBC News 2005, World Edition 2005a. mental risk as a consequence of the consumption patterns of protected spring, rainwater collection. Improved water sources do not
higher income groups and the production and distribution sys- include: unprotected well, unprotected spring, rivers or ponds, vendor-
tems that serve them. The pollution and contamination pro- provided water, bottled water (due to limitations in the potential quantity,
duced by higher-income groups are immediately felt by lower- not quality, of the water), tanker truck water. Access to improved
■ Poor solid waste management 250,000 people, has become the city’s de facto dump: approxi- income groups who live and work in areas and industries that sanitation facilities refers to the percentage of the population with access
mately 1,600 tons of solid waste is dumped every day on land absorb it. The poor consequently bear heavy health burdens to: facilities connected to a public sewer or a septic system, pour-flush
Less commonly researched than sanitation provision is a formerly intended for housing. The sprawling waste has been and other barriers to escaping poverty. latrines, simple pit or ventilated improved pit latrines. (United Nations
related challenge to sustainability: the dearth of adequate solid blamed for numerous illnesses among the residents and has Low-quality, overcrowded housing and lack of basic services Department of Economic and Social Affairs, Statistics Division 2005.)
waste management in many cities around the world. contaminated a nearby spring, upon which many residents often lead to illness and absenteeism, which in turn affect the 26 See Part Two of this report for an in-depth discussion on water and
Municipalities in developing countries commonly spend 20 to depend for water when the municipal supplies stop working.33 economic growth and sustainability of the urban system as a sanitation issues in cities.
50 per cent of their available budgets on solid waste manage- Thin plastic carrier bags given out by supermarkets and local whole. Even as the consumption patterns of higher-income 27 Ray 2003.
ment, but many are only able to collect 30 to 60 per cent of the vendors comprise much of the solid waste in Dandora and sim- groups lead to suffering for the lowest-income residents, each 28 UNESCO 2003.
waste in their cities while serving just half of the population. ilar neighbourhoods. Kenya produces 48 million of the bags each depends on the other, for labour, employment and the avail- 29 Thompson, et al. 2000.
In Nairobi, Kenya, for example, only 25 per cent of the city’s year, and businesses in Nairobi hand out more than 2 million of ability of city services. Improving the urban environment in 30 UNESCO 2003.
daily waste is collected.32 Where sanitary landfills, recycling them annually. The bags can take up to 1,000 years to decom- 31 Ibid.
ways that remediate the harms experienced by the urban poor
programmes and other properly managed means of solid waste pose; along with dumps and landfills, many bags end up in 32 BBC News World Edition, 2005b.
can reduce poverty and increase the overall economic, environ-
33 Kantai 2003.
disposal are not available, open dumping and burning are the drains, sewers, riverbeds and the sea. Nobel Peace Prize laureate mental and social sustainability of cities. Making progress
34 BBC News World Edition, 2005b.
norm. This leads to environmental and health hazards for Wangari Maathai has also linked Kenya’s plastic bag pollution to toward sustainability, therefore, means designing and improv-
35 Atkinson 1996.
urban residents – especially those who live closest to the dump the incidence of malaria: discarded plastic bags trap rainwater, ing urban systems with people in mind – particularly those
sites. The Nairobi neighbourhood of Dandora, home to providing ideal breeding grounds for mosquitoes.34 with the fewest resources.

134 135
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
3.7 Double Jeopardy:
H
uman conflicts and natural disasters impact
cities differently – and often more deeply –
than rural areas. Conflicts lead to the growth
and proliferation of slums as displaced people
The Impact of Conflict and Natural Disaster on Cities seek refuge at the margins of urban areas;
buildings and roads crumble and fall in the wake of major
tremors, landslides and floods. The sheer concentration of peo-
ple and infrastructure in cities often means greater loss of life
when disaster strikes, and the social, political and structural
capacity of cities to provide shelter for those in need is often
limited. When conflicts or natural disasters hit, they can wreak
havoc on urban economies, destroy communities and tear fam-
ilies apart. Such events perpetuate urban poverty, placing addi-
tional strains on people and places already burdened by lack of
resources.
■ Conflicts generate slums
Conflicts and crises in war-torn countries often result in the
mass exodus of rural communities to urban areas, where most
end up in low-income, poorly serviced settlements, or slums.
The continued threat of conflict in countries is, therefore, a sig-
nificant contributing factor in the proliferation of slums in
urban areas.1 In Sudan, for instance, urban areas accommodat-
ed two-thirds of the more than 6 million internally displaced
persons (IDPs) in the country in 1998; almost half of these
IDPs moved to the capital city, Khartoum. Surveys indicate
that the majority of IDPs in Khartoum are from Southern
Sudan, the region most affected by a protracted civil war, and
most reside in squatter settlements on the periphery of the city,
with little access to basic services.2
The plight of displaced persons, whether they are settled in
large groups in camps or merged into urban slums, therefore,
Collapsed buildings after an earthquake in the Taiwan Province of China LO TSUNG HSIEN/UNEP/STILL PICTURES
needs special attention. The urban context presents unique
issues and dynamics for people uprooted from their homes,
separated from their families and community networks, and
stripped of their livelihoods by conflict. Urban IDPs may come
from different areas of the country, different post-conflict situ-
ations and possibly from different sides of the conflict, making
resettlement in a diverse environment difficult.
In Azerbaijan, where conflicts with neighbouring Armenia
have raged since 1988, the total number of internally displaced
persons stands at nearly at 1 million.3 Approximately 40 per
cent of the country’s displaced population lives in urban areas,
which have proven unsuitable and unacceptable for long-term
habitation, especially when employment opportunities are
scarce. Almost 95 per cent of these urban IDPs state that they
wish to return to their former homes – not because the materi-
al living conditions in rural settlements are better, but because
they want to be able to continue engaging in their former
source of livelihood and continue living in a familiar environ-
ment with their own community networks. Chronic insecurity
or lack of rehabilitation of disaster-struck areas often inhibits
the return process, leaving people stranded in substandard liv-
ing conditions in urban areas.
136
In some countries, however, refugees and IDPs residing in
urban areas often have no intention of returning to their place
of origin, as the movement to cities is seen as an opportunity to
escape the impoverishment of their rural homes. A review of
rehabilitation and reconstruction programmes in Afghanistan
conducted in the 1990s showed that a significant number of
IDPs in cities such as Kabul and Mazar-i-Sharif used the
opportunities presented by disruption to leave their rural
homes for good and most had no intention of returning.4 In
some war-torn countries, IDPs literally have no place to return
to as their homes and lands have been taken over forcefully by
warring factions. This is the case in Somalia where years of law-
lessness has led to illegal occupation of rural and urban land by
armed militias.
City governments and municipalities have to deal with the
burden of increased pressure on urban infrastructure caused by
an influx of refugees and IDPs. Most IDPs and refugees are at
risk and in need; they often live in densely populated squatter
settlements on the periphery of cities, where widespread pover-
ty and underdevelopment are prevalent. Local authorities face
considerable additional pressure to absorb large numbers of
refugees and internally displaced persons during the conflict; in
many cases, they even have to deal with IDPs and refugees long
after the crisis is over. For instance, preliminary reports indicate
that despite the signing of a peace accord in January 2005,
many of the IDPs in Sudan’s urban areas are unwilling to return
to their rural homes, either because assistance to reintegrate
into their local communities is insufficient or local communi-
ties and administrations are acutely under-resourced to manage
the return process.5 Some IDPs have tried to return to their
areas of origin only to find inadequate resources and services,
forcing many to return to urban centers in an effort seek oppor-
tunities for a new start.
■ Low-income settlements are more vulnerable to
natural disasters
Urban settlements are also prone to threats from natural and
environmental hazards, and people living in poverty every-
where, especially in urban areas, are most at risk. Substandard
housing and construction practices, lack of infrastructure,
absence of secure tenure, inappropriate land use and increasing-
ly degraded environments leave large sections of the poorest
communities chronically vulnerable.
The world’s largest cities are concentrated in developing
countries, and many of them are in areas where earthquakes,
floods, landslides and other disasters are most likely to happen.
According to the UN’s Bureau for Crisis Prevention and
Recovery, some 75 per cent of the world’s population lives in
areas that were affected at least once by an earthquake, a tropi-
cal cyclone, floods, or drought between 1980 and 2000.6
At the same time, natural and man-made disasters are increas-
ing in regularity, and perhaps more importantly, their adverse
impacts on populations and human settlements are rising.
137
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
War-torn Luanda, Angola EDUARDO LÓPEZ MORENO
Calculations by the International Federation of the Red Cross
and Red Crescent show that from 1994 to 1998, reported dis-
asters averaged 428 per year. From 1999 to 2003, this figure
shot up by two-thirds to an average of 707 natural disasters per
year. The sharpest rise occurred in developing countries, which
suffered an increase of 142 per cent.7
Poor people in developing countries are particularly vulnera-
ble to disasters because of where they live; they are more likely
to occupy dangerous floodplains, river banks, steep slopes and
reclaimed land, and their housing is less likely to survive a
major disaster. For instance, in Latin America, hundreds of
low-income urban dwellers lost their lives when their precari-
ously situated homes were swept away in floods and landslides
during Hurricane Mitch in 1998. And investigation into the
2003 earthquake in Bam, Iran, found that most of the 40,000
people killed lived in housing that was built in the traditional
mud-brick style without the necessary supportive structures to
withstand tremors.8
In poor countries, the impact of disasters can be devastating
on the economy; disasters not only wipe out decades of devel-
opment in a matter of hours, but they also have a significant
impact on a country’s gross domestic product (GDP). Figures
138
Conflicts and crises in war-torn countries
often result in the mass exodus of rural
communities to urban areas, where most
end up in low-income, poorly serviced
settlements, or slums.
compiled by the World Bank show that between 1990 and
2000, natural disasters resulted in damages constituting
between 2 and 15 per cent of the affected countries’ GDP.
While industrialized countries suffer higher losses in dollar
terms – mainly because the cost of repairing or replacing
destroyed infrastructure is higher – the overall impact of disas-
ters on the economies of rich countries is negligible. According
to the International Federation of the Red Cross and Red
Crescent Societies, disasters in industrialized countries have
inflicted an average damage of $318 million per event, com-
pared with $28 million per event in developing countries.
However, industrialized countries are able to quickly recover
from the impact of disasters, mainly because of a surge in
reconstruction activities and more public spending on rehabil-
itation of the affected areas. Moreover, rich countries are often
more prepared to deal with the consequences of disaster, as they
have more medical and emergency assistance services than
lower-income nations. The prevalence of life and property
insurance in the developed countries also means that affected
populations suffer less personal financial loss than their coun-
terparts in developing countries.
Disasters can paralyse developing countries, or even perma-
nently destroy their social and economic assets. In Aceh,
Indonesia, for instance, the total estimate of damage and losses
from the December 2004 tsunami was $4.45 billion – nearly
97 per cent of the region’s GDP.9 Many developing countries
also lack the health facilities to deal with large numbers of
injured patients, resulting in a higher eventual death toll than
in countries better equipped for disaster. The United Nations
Office for the Coordination of Humanitarian Affairs (OCHA)
estimates that in the 1990s, natural disasters killed almost seven
times more people in developing countries per event than in
industrialized countries; an average of 44 people per event died
in industrialized countries compared with 300 people per event
in developing counties.10
Surviving Anarchy: Somalia’s Experience
Somalia UNICEF
Human settlements in Somalia have been
severely affected by more than a decade of
civil war, which has not only caused the
destruction of infrastructure and services, but
has also led to the breakdown of government
institutions.The country had no functioning
government for almost 14 years as warlords
controlled most regions. In many cities, local
and national government records have been
ravaged and capacities have been greatly
reduced or are non-existent.This has made it
difficult to measure the country’s level of
human development or to make an assess-
ment of needs vis-à-vis the Millennium
Development Goals.
Although a new interim government was
finally elected and installed at the end of
2004, the effects of the conflict will take years
to erase, as much of the country has no fully
functioning governance structures, either at
the national or local government level. Nine
months after the elections, for instance, the
new government had still not moved to the
capital Mogadishu – which is still considered
to be too dangerous – and was operating
from Jowhar, a town north-east of the former
capital city.The elections themselves were
conducted in neighbouring Kenya, which is
host to thousands of Somali refugees and
exiles.
Somalia is categorized as one of the 27 high-
priority countries for the achievement of the
Millennium Development Goals; available sta-
Sources: UN-HABITAT 2005b; Hargeisa Municipality 2003; BBC News 2005a.
tistics indicate that the maternal mortality
ratio per 100,000 births is 1,600 – much high-
er than that of neighbouring countries such
as Kenya (590) and Eritrea (1000); primary
education enrolment is only 13.6 per cent.The
lack of a fully functioning government and
the disruption caused by conflict makes the
achievement of social and economic develop-
ment precarious.
Despite the anarchic situation, since the early
1990s, isolated communities within Somalia
have been able to achieve some level of sta-
bility and even manage to govern themselves
despite the absence of a legitimate national
government. In May 1991, five months after
the state of Somalia collapsed and fell into
civil war, the north-west region claimed inde-
pendence and formed its own government
with its own currency and institutions.
Although it was not internationally recog-
nized as an autonomous state, the Republic of
Somaliland developed both formal and infor-
mal local governance structures, and even
held its first local elections in 2002 and multi-
party presidential elections in 2003.
In a country with no formal banking institu-
tions, a widespread system of money vendors
developed, allowing local and international
agencies to make financial transactions.The
relative stability in Somaliland ensured that
the region invested in its own development
and also attracted foreign capital and assis-
tance. Compared to other regions in Somalia,
Somaliland not only has more hospitals, but
also has a vibrant private sector, which gener-
ates revenue for the region.The economy is
also highly dependent on money sent home
by members of the Somali diaspora. Although
the economy of Somaliland is dominated by
trade in livestock, the service sector, including
mobile phone companies, has shown positive
results in recent years. In the capital city,
Hargeisa, revenue collection has become
more efficient and the city of more than
400,000 inhabitants has become the centre of
international aid agencies’ operations as inse-
curity in other regions only allows limited
operations.
However, like many war-torn cities, Hargeisa is
becoming a destination for returnees and
internally displaced persons (IDPs).The
Hargeisa Municipality Statistical Abstract,
published in 2003 by the Municipality of
Hargeisa, shows that while the city has per-
formed relatively well compared to other
cities in the region and in the rest of Somalia,
the spontaneous growth of settlements com-
prising returnees and IDPs is making it diffi-
cult for the municipality to provide basic serv-
ices. In fact, in the city as a whole, only one out
of seven dwellings has access to piped water
and 20 per cent of households live in tempo-
rary, makeshift structures.
With an annual urban growth rate of 5.7 per
cent, Somalia is one of the most rapidly
urbanizing countries in Eastern Africa.The
transition to democracy is likely to increase
rates of urbanization, as returnees and IDPs
make their way to cities. Land disputes are
also likely to come to the fore, as years of law-
lessness led to illegal occupation of land by
militias.
Although Somaliland has distanced itself from
Somalia’s new transitional government –
which it perceives as a threat to Somaliland’s
autonomy – this region of 3.5 million people
has already begun working to ensure that
urban governance structures are strength-
ened as national priorities are being rede-
fined.The Municipality of Hargeisa is working
with other local authorities to institute urban
land reforms through the formulation of a
City Charter on a range of issues, such as plan-
ning and taxation. Funded by UNDP and
implemented by UN-HABITAT, the project’s
aim is to increase efficiency in urban planning,
management and development, as well as
service delivery and fiscal management.
UN-HABITAT is also working with the three
regional authorities in Somalia, namely
Somaliland, Puntland and South-Central
Somalia, to improve the capacities of local
authorities and civil society in basic leader-
ship and urban planning – a need that was
identified by the 2002 Somalia Urban Sector
Profile Study.The European Commission,
UNDP, UNICEF and the Governments of Italy
and Japan are working with UN-HABITAT to
implement projects in these areas. However,
the success of these initiatives will largely
depend on how effective the new govern-
ment is in securing lasting peace in a country
that has been managed through fear and vio-
lence for more than a decade.
139
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

difficult aspects of the post-crisis recovery process, yet many local

resources can be tapped to assist with economic recovery, includ-
ing the technical knowledge of skilled and semi-skilled local
workers, the eagerness and assets of entire communities, and the
resources of local authorities and the private sector.
Understanding urban vulnerability is the first step toward
developing mitigation strategies that effectively improve
resilience and reduce vulnerabilities of urban populations in the
long term. The cornerstone of the implementation strategy is to
build a “culture of prevention”, or disaster mitigation, among
the society at large. Disaster mitigation not only saves lives but
also makes economic sense. The World Bank and the U.S.
Geological Survey estimate that economic losses worldwide
from natural disasters in the 1990s could have been reduced by
$280 billion if $40 billion had been invested in preventive
measures.13 In China, the World Bank estimated that the $3.15
billion spent on flood control since the 1960s has averted losses
Earthquake victims in Kashmir UNHCR/V. TAN
Endnotes
1 It is interesting to note that the region with the most conflicts - sub-
Saharan Africa - also has the highest proportion of its urban residents
living in slums.
2 Eltayeb 2003.
3 This comprises roughly 13 per cent of the country’s population, which is
one of the highest such rates in the world. See UNDP’s Azerbaijan
Human Development Report 2000.
4 This review was jointly conducted by UNCHS and UNDP in 1995.
5 See www.idpproject.org/sudan for more information.
6 United Nations Office for the Coordination of Humanitarian Affairs
2005a.
7 Ibid.
8 United Nations Office for the Coordination of Humanitarian Affairs
2005a.

The year 2005 was a particularly costly one in terms of lives
lost and damage inflicted by natural disasters around the world.
Nearly 125 million people were injured, lost their home, or
required other immediate assistance as a result of disasters that
year. More than 100,000 people were killed, in addition to the
230,000 who died in the tsunami at the end of 2004. Total eco-
nomic damages in 2005 reached a record $200 billion, includ-
ing $125 billion in losses from Hurricane Katrina alone. The
single greatest human toll followed the October earthquake in
Pakistan and India, the repercussions of which continued for
months as affected families weathered out a difficult winter in
makeshift shelters.11
■ Sustainable recovery from crisis
Disasters have serious consequences at every level, from far-
reaching economic losses to personal hardship for individual
families. The broad impacts of disasters exacerbate the funda-
mental challenges of crisis management and recovery processes:
how to bridge the gaps that have repeatedly emerged between
emergency recovery and sustainable development efforts, and
how to provide all stakeholders with practical strategies to miti-
gate and recover from crises. The concept of sustainable recov-
ery12 does not entail an abrupt shift from relief to development,
but rather an integrated approach in which those involved
attend to basic needs while also supporting longer-term sustain-
able development.
Conflicts and disasters perpetuate poverty by placing an addi-
tional strain on already precarious social, environmental and
economic conditions. Persistent urban poverty and lack of
resources again increase vulnerability, weaken coping strategies
and delay the recovery process. The urban poor are forced to
accept a greater degree of risk because they lack the resources to
live or work in safer environments. Urban poverty alleviation
must therefore be central to any plan to effectively manage
urban disasters and to sustain peace and stability. Other crucial
pillars of sustainable recovery are good governance, public par-
ticipation, inclusive decision-making, institutional development
and empowerment of civil society. When governments adopt
policies to make livelihoods more secure, institutions more
responsive, public-private partnerships more effective, commu-
nities more safe and sustainable, and poverty less prevalent, per-
sonal and social protection are dramatically enhanced.
The need for durable settlement solutions for internally dis-
placed populations is one of the key issues in post-crisis urban-
ization. In practical terms, this means either helping displaced
people resettle in their areas of origin, or aiding their effective
and sustainable social, economic, legal, and political integration
into urban communities. Either way, the importance of support-
ing greater self-reliance among the displaced is apparent, in par-
ticular by ensuring their access to land, income-generating activ-
ities and skills development. The emergency phase after disaster
or conflict tends to frame displaced populations as beneficiaries
rather than partners in the process and agents of development.
Economic recovery, for example, is recognized as one of the most
of about $12 billion. Similarly, more federal funding for the lev- 9 UNEP 2005.
ees in New Orleans might have reduced the scale of the tragedy 10 United Nations Office for the Coordination of Humanitarian Affairs
when Hurricane Katrina struck in August 2005. 2005b.
Medellin, Colombia, provides a good example of successful 11 Worldwatch Institute 2006a.
community-based disaster prevention. In the mid-1980s, fol- 12 See UN-HABITAT Sustainable Recovery and Reconstruction framework,
lowing the destruction of the city of Armero by mudslides trig- www.unhabitat.org.
gered by a volcanic eruption, the Colombian government 13 Figures cited in IRIN News, United Nations Office for the Coordination
established a National System for Disaster Prevention and of Humanitarian Affairs (OCHA) 2005b.
Response. When a major landslide struck Medellin in 1987, 14 Medellin is Colombia’s second largest city with 2 million inhabitants,
the city and its inhabitants were able to mobilize resources to close to 10 per cent of whom live in informal settlements on steep
create a safer living environment, integrating risk management hillsides, vulnerable to floods and landslides. The landslide of 1987
strategies with municipal physical, social and economic plan- killed more than 500 and left 3,500 inhabitants homeless. (Boulle &
ning. Thanks to combined civic education and political and Palm 2004.)
financial commitment, the landslides in Medellin decreased
from 533 in 1993 to 191 in 1995.14 Vulnerability reduction
plans and disaster risk considerations are ideally integrated into
sustainable development policies, planning and programming
– in particular at local levels.
As the nature of disasters in cities becomes more multifaceted,
so must the approach to their management. The impact of the
recent Indian Ocean tsunami is a tragic reminder of the extreme
vulnerability of the built environment to natural hazards.
Natural disasters in and around cities are often anything but
“natural”, being triggered by deficient urban management prac-
tices, inadequate planning, excessive population densities, eco-
logical imbalance, inadequate investments in infrastructure, and
poorly prepared local governments. Furthermore, the increasing
number of people displaced by crises and seeking refuge in cities
is a call for attention all slum dwellers deserve: to improve their
living conditions and address the urban context of poverty, as
spelled out in the Millennium Development Goals.
Paradoxically, a crisis can also be an opportunity. During
recovery from a disaster, communities have a unique oppor-
tunity to revisit past practices and rewrite policies to affect
future development. In Rwanda, for instance, new land laws
were instituted after the genocide in 1994 to give women and
other vulnerable groups more rights to inherit and own land
and property.

140 141
 S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
New Orleans: Poor Residents Suffer Deepest Impact
of Hurricane Katrina
When Hurricane Katrina struck the Gulf
Coast of the United States on 29 August
2005, the storm left more than one million
people homeless and killed hundreds
across three states. The city of New Orleans,
in the southern state of Louisiana, suffered
Katrina’s greatest lasting impact.
Lashed by winds of more than 140 miles per
hour and flooded by water overflowing the
levees that kept the Mississippi River, the
Gulf of Mexico and Lake Ponchartrain at bay,
New Orleans lay almost entirely submerged
and in ruins after the storm. Many of the
city’s 485,000 inhabitants fled before the
storm via the well-planned interstate high-
way evacuation route. For tens of thousands
of people without cars, cash or anywhere
else to go, however, Mayor C. Ray Nagin’s
evacuation order had meant little.The
refugees, disproportionately African
American and living in poverty, took shelter
under freeway bridges, in the city’s sports
arena, and in the nearby convention centre
for days until help arrived. Reporters cover-
ing the scene likened it to something more
akin to war-torn Somalia or post-tsunami
Indonesia than a scene from one of the
world’s wealthiest nations.The disaster fast
became a symbol of race and class division
in the country.
Before Katrina pummeled its shore, the
below-sea-level city of New Orleans was
known for its European charm and grand-
scale urban fêtes, including one of the
largest Mardi Gras celebrations in the
world. With the storm came a flood of
Aftermath of Hurricane Katrina, New Orleans September 3, 2005 DIGITAL GLOBE
facts about real-life conditions for average
denizens of New Orleans, nearly one-quar-
ter of whom were elderly or disabled, and
more than 28 per cent of whom lived in
poverty – double the national average of
12.4 per cent. Of those living in poverty, 84
per cent were African American and 43
per cent were children under the age of 5.
Access to a car, the primary means out of
the city during the crisis, was equally dis-
proportionate: among African American
households, 35 per cent did not have a car,
Sources: Applebome, et al. 2005; Greater New Orleans Community Data Center 2005; DeParle 2005; Leavitt 2000; Teather 2005; Gonzalez 2005; Walsh, et al. 2005; Luthra 2005.
142
while the same was true of only 15 per
cent of white households.
As in many parts of the developing world,
the poorest residents of New Orleans lived
in the most hazardous areas of the city.
Many of the city’s lowest-income residents
lived in the floodplains of the Lower Ninth
Ward, a neighbourhood that sat below sea
level and was inundated when the canals
and levees failed. Although the Federal
Emergency Management Agency had pre-
dicted that a hurricane would strike New
Orleans since at least 2001, federal funds
to reinforce the levees had been decreas-
ing in recent years.
Speaking to a New York Times reporter,
geographer Craig E. Colten of Louisiana
State University said,“Out West, there is a
saying that water flows to money. But in
New Orleans, water flows away from
money. Those with resources who control
where the drainage goes have always cho-
sen to live on the high ground. So the peo-
ple in the low areas were hardest hit.”
The Lower Ninth Ward neighbourhood –
where more than 98 per cent of the residents
were African American and more than a third
lived in poverty – was built on a reclaimed
cypress swamp, gradually drained and devel-
oped over the first half of the 20th century.
The city’s higher ground had been settled
since the early 1700s, when French colonists
fortified the swampland surrounded by large
bodies of water and called it “leflontant”–
the floating island. By 2005, the New Orleans
metropolitan area was home to more than
one million people.With population expan-
sion came more reclamation; levees were
built and water pumped away as settlements
spread down from the high southern shore
of Lake Ponchartrain to the low banks of the
Mississippi.The lowest land was the only
place European immigrants and African
American families could afford to build
homes in the early 1900s; the dirty, flood-
prone parcels of land were adjacent to the
city’s commercial and industrial areas.
Poverty kept residents of the Lower Ninth in
place over the years, unable either to move
up and out or to renovate their increasingly
run-down houses. In 2000, more than half of
the neighbourhood’s residents owned their
homes and had occupied them for 10 years
or longer; the opposite was true nationwide,
with more than 60 per cent of American
households having moved in the past 10
years.The houses in the neighbourhood
held the history of New Orleans itself – 62
per cent were built before 1960, and only
one-tenth were less than 20 years old.The
age of the housing, along with its location,
put residents of the Lower Ninth at risk.
Having endured flooding before, the resi-
dents suffered another deep impact on their
stability and their access to affordable hous-
ing with Hurricane Katrina.They may also be
the last to benefit as the city is gradually
rebuilt over the next several years.
Katrina was the first such storm to devas-
tate a major urban centre in the United
States. In other places similarly affected,
the rebuilding of affordable housing has
historically taken last place among the
items on long-term plans, or has been left
out altogether. Kobe, Japan, provides a
case in point. When that city was
destroyed by an earthquake in 1995, many
residents lived in temporary housing for
eight years, and areas of the city that had
been affordable for families were rebuilt
with housing beyond their financial reach.
But while the city of New Orleans will no
doubt be eventually rebuilt, many fear its
soul has been lost forever. As neighbouring
cities and states struggled to cope with the
hundreds of thousands of refugees fleeing
the flooded city, authorities warned that
draining the water from the city could take
months, which would make it less likely that
the refugees would return soon. According
to one report, many of the more than
200,000 people who crossed into the neigh-
bouring State of Texas in buses, planes and
trains vowed never to return to New
Orleans and its surrounding areas.
143
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
3.8 Urban Insecurity: New Threats, Old Fears
■ Cities as targets
P
rior to 11 September 2001,1 urban violence and
insecurity were considered peripheral to the devel-
opment concerns of both rich and poor countries.
In recent years, however, poverty, under-develop-
ment and fragile states have created fertile condi-
tions for the emergence of new threats, such as transnational
crime and international terrorism, which are being played out
in the world’s cities.
Recent attacks on New York, Washington, Madrid, London,
Nairobi, Dar es Salaam and Bali, among other cities, have
demonstrated that urban insecurity is an emerging internation-
al issue that impedes economic growth and sustainability.
Although cities have been the sites of warfare in both pre-mod-
ern and modern times, the near-invisible nature of modern-day
attackers has made urban warfare and terrorism much more
difficult to track and control, and is having a far more devastat-
ing financial, physical and psychological impact on cities than
ever before. New York City is estimated to have lost $110 bil-
lion in infrastructure, buildings, jobs, and other assets as a
result of the 11 September attacks.2 The impact of the attacks
has also extended outside the city’s boundaries. The World
Bank estimates that as a result of the terrorist attacks on New
York and Washington, global gross domestic product (GDP)
was reduced by 0.8 per cent, and some 10 million additional
people were added to the world’s poor.3
Unfortunately, the “war on terror” threatens to sideline the
struggle against poverty. The security measures adopted by
European and North American cities shortly after the recent
attacks have greatly compounded terrorism’s effects on poverty
reduction. Increased security measures, such as metal detectors,
street surveillance, stricter control of public areas and, in certain
cases, curtailment of civil liberties, threaten the essence of cities.4
Recent estimates indicate that the combined total expenditure on
programmes for improving access to clean water, building sewage
systems, reducing hunger, preventing soil erosion, eradicating
illiteracy, immunizing children, providing reproductive health
care to women, and fighting HIV/AIDS and malaria added up
to a little more than half of the military budget appropriated for
the Iraq war by the United States in 2004.5
Cities are not only targets of international terrorism, but also
of localized ethnic and religious conflicts. According to the
Worldwatch Institute, urban unrest is likely to increase in the
largest cities of the developing world as more and more people
from diverse ethnic and religious groups increasingly come into
close contact with each other.6 Although cities offer the oppor-
144
tunity for diverse interests to integrate, when resources are
scarce, or when political interests collide, they can become the
sites of warfare. In India, for instance, centuries-old animosities
and grievances were played out in the country’s biggest cities in
1992 when Hindu militants descended on the small town of
Ayodhya, attacking security forces and destroying a 16th centu-
ry mosque. Rather than spreading through the nearby country-
side, the hatred exploded hundreds of kilometres away, in
Mumbai, Calcutta, Ahmedabad, and New Delhi. In total, 95
per cent of those killed in the communal riots that ensued were
city dwellers.7 Cities such as Los Angeles, Belfast, Sarajevo and
Mogadishu have all suffered from one form of urban warfare or
another, creating “a new phase in the life of cities, where the
concentration of ethnic populations, the availability of heavy
weaponry, and the crowded conditions of civic life create futur-
istic forms of warfare … and where a general desolation of the
national and global landscape has transposed many bizarre
racial, religious, and linguistic enmities into scenarios of unre-
lieved urban terror”.8
■ Urban crime
Evidence shows that the probability of being a victim of crime
and violence is substantially higher in urban areas than in rural
areas. Approximately 60 per cent of urban dwellers in Europe
and North America and 70 per cent of urban dwellers in Latin
America and Africa have been victimized by crime over the past
five years.9
Overall, recorded crime rates are stabilizing or even decreasing
in some countries, but the risk of being a victim of a violent
crime such as homicide, assault, rape, sexual abuse, or domestic
violence has continued to rise worldwide. Globally, more than
1.6 million people die as a result of violence every year.10 The
increasing availability and use of firearms lends heavily to the
increase in urban violence. On average, violence makes up at
least 25 to 30 per cent of urban crime11 and women, especially
in developing countries, are twice as likely to be victims of vio-
lent aggression (including domestic violence) as men.12
Increases in violence against women can be correlated with
declining household economic security, but it is clear that
poverty and unemployment on their own do not cause crime,
violence and abuse. Rather, the costs of poverty and unemploy-
ment in the form of stress, loss of self-esteem and frustration
appear to influence violent behaviour.13
Urban insecurity presents a major challenge to the social and
economic development of cities because it compounds other
factors, such as poverty and social exclusion, which already limit
the quality of life for many. Violence and crime are no longer
viewed exclusively as criminal problems but also as problems
affecting the development of societies. Insecurity contributes to
the isolation of groups and to the stigmatization of neighbour-
hoods, particularly those in which the poor and more vulnera-
ble live. It creates conditions of fear, hinders mobility and may
be a major stumbling block for participation, social cohesion
and full citizenship. The most excluded groups – women, chil-
dren, the elderly, widows, and people living with HIV/AIDS –
are typically cut off from networks that provide access to power
and resources, making them more vulnerable and increasing
their risk of remaining poor or sinking further into poverty.14
Communities where an increasing proportion of the population
is excluded from society also suffer from higher levels of crime
and violence than those that are connected to mainstream net-
works and power structures. Victimization surveys conducted
by UN-HABITAT and its partners have also shown that people
living in poverty are more likely to be victims of crime than
higher-income residents.
Poverty is often cited as a cause of crime and violence, but
increasing evidence suggests that poverty per se has little to do
with crime and violence levels; rather, crime and violence
occur more frequently in settings where there is an unequal
distribution of scarce resources or power coupled with weak
World Trade Center site © NATHAN GOODMAN. IMAGE FROM BIGSTOCKPHOTO.COM
Poverty, underdevelopment and fragile
states have created fertile conditions for
the emergence of new threats, such as
transnational crime and international
terrorism, which are being played out in
the world’s cities.
institutional controls. Crime is often linked to institutional
weaknesses in society. Crime increases when the social control
that operates through formal institutions – such as the police
and judicial systems – and informal institutions, including
civil society organizations and solidarity networks, breaks
down or is weakened. Although there is no simple or direct
causal relationship between inequality and violence, inequali-
ty does appear to exacerbate the likelihood of violent crime,
especially when it coincides with other factors. For instance,
Africa and Latin America – the regions with the highest levels
of income inequality – exhibit high levels of homicide, which
is often used as a proxy for the broader category of violent
crime.15 Cities where inequalities are most stark also appear
more vulnerable to insecurity. In 2005, for instance, South
Africa reported over 800 protests in the nation’s slums, many
of which turned violent.16 Some believe that this can be
explained by the theory of “relative deprivation”, that is,
145
 S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Manila FRIEDRICH STARK / STILL PICTURES
Worldwide, the majority of criminal offences are committed by youth
between the ages of 12 and 25, and recently, youth delinquency has
become increasingly violent.
inequality breeds social tension as those who are less well-off
feel dispossessed when comparing themselves to others. This
theory is based on the assumption that individuals or groups
are more likely to engage in violence if they perceive a gap
between what they have and what they believe they deserve.
The consequences of relative deprivation seem to be playing
out in the world’s cities, which are sites of extreme inequality.
This is not the case in rural areas, where levels of deprivation
or prosperity are likely to be more evenly distributed.
Crime tends to impact people living in poverty more deeply
and intensely than it does higher-income residents. Not only are
low-income people often unable to protect themselves from
crime, which can heighten their sense of helplessness and pow-
erlessness, but they also lack adequate fall-back systems, such as
insurance and savings, making recovery from the psychological
and material impacts of crime difficult. Replacing stolen goods
146
– such as the bicycle used to get to work – may be impossible,
leading to further hardship. Inequality and exclusion exacerbate
insecurity, which perpetuates the vicious cycle of poverty and
vulnerability. Surveys conducted by UN-HABITAT in Nairobi,
Johannesburg and other cities indicate that people living in
poverty cite safety and security as a major concern – as impor-
tant as hunger, unemployment and lack of safe drinking water.
Supporting the physical security of the lowest-income urban res-
idents is therefore crucial to reducing poverty.
■ Youth, unemployment and crime
Worldwide, the majority of criminal offences are committed
by youth between the ages of 12 and 25, and recently, youth
delinquency has become increasingly violent. Youth unemploy-
ment is a significant contributing factor, given the high propor-
tion of young people, high rates of population growth and slow
economic growth in many cities. Some studies have also sug-
gested a link between excessively high urban growth rates and
violence and conflict in cities.17 According to Population
Action International, countries with rapid rates of urban pop-
ulation growth – greater than 4 per cent per year – were rough-
ly twice as likely to experience civil conflict during the 1990s.18
Unemployment tends to be two or three times higher for
young people than for the general population, and the lack of
work opportunities may increase frustration, especially if young
people’s expectations have been raised through expansions in
education. Estimations for Africa reveal that more than 8 mil-
lion people enter the labour market each year for whom jobs
will have to be found.19 In developed countries, youth unem-
ployment is usually twice the rate of adult unemployment; in
developing countries it is often much higher. According to the
International Labour Organization (ILO), an estimated 88 mil-
lion young people between the ages of 15 and 24 were without
work in 2003, accounting for nearly half the world’s jobless. In
the developing world – home to 85 per cent of youth – unem-
ployment in this group is particularly high.
The incapacity of a country to integrate a young labour force
into the formal economy has a profound impact on the coun-
try as a whole, ranging from the rapid growth of the informal
economy to increased national instability. But while the infor-
mal sector offers a solution to urban unemployment, it is char-
acterized by low salaries, dangerous work and job insecurity, all
of which make it harder for youth to escape poverty.
Long-term unemployment among youth is known to be
associated with negative consequences such as ill health,
involvement in crime and delinquency and substance abuse.20
In this context, the boundary between what is legal or lawful
and what is illegal and illicit becomes ambiguous.
Disenchanted urban youth are among the first recruits to
organized criminal gangs and violent rebel groups. However,
urban conflict and unrest is not simply confined to the poor.
Studies have shown that the risks of instability among youth
may increase when skilled members of higher income and
social groups are marginalized due to lack of opportunities or
when the salary or benefits they receive are not commensurate
with their socioeconomic background or educational achieve-
ments, and hence what they feel they are entitled to earn.21
Unemployed youth seek alternative models of success and peer
recognition, which sometimes implies illicit and criminal activ-
ities but may also lead to violent behaviour.22
■ Fortress cities and the architecture of fear
High levels of urban crime and violence also impact the
social fabric of entire cities; they instil fear and suspicion in the
lives of urban residents, often leading to residential fortification
among the rich, who build higher walls around their homes
and spend more on private security, in effect “locking them-
selves” in enclaves that are physically separated from the rest of
the city.23 The fear of crime has led to increased fragmentation
and polarization of urban communities, characterized by
enforced segregation through gated communities, stigmatiza-
tion and exclusion.24 Insecurity has resulted in the abandon-
ment and stigmatization of certain neighbourhoods and the
development of an architecture of fear and the gradual establish-
ment of so-called “fortress cities” where response to crime has
led to spatial transformation that has changed parts of cities
into protected enclaves and “no-go areas” separated by high
walls, gates, electronic surveillance cameras and private securi-
ty guards. As one commentator put it, “Creating fortified envi-
ronments may reduce the opportunities for crime but may raise
levels of fear.”25 The result is a fragmented urban environment
that may contribute to the fear of crime outside protected areas,
which could make cities more vulnerable in the long term.
In cities of the developed world, new forms of international
terrorism that target public infrastructure, such as under-
ground train networks, have promoted a culture of fear among
urban residents. In some countries, such as the United
Kingdom, the threat posed by international terrorism has
resulted in stringent immigration policies and stricter policing,
which threaten to polarize urban communities even further.
There are fears that the threats posed by international terror-
ism may also lead to new forms of xenophobia in European
and North American cities.
Dealing with perceptions of crime, particularly anxiety and
fear of crime, is as important as reducing crime levels. Fear of
crime affects quality of life and has negative economic and
political consequences. It can also affect people’s willingness to
trust, interact and cooperate with the authorities, particularly
the police, but also with local government crime prevention
practitioners.26 Fear of crime does not affect everyone to the
same extent. The most vulnerable in society, such as women,
the elderly and the poor, fear crime the most and have the
most difficulty recovering from it.27
■ Crime makes cities less competitive
Crime and lawlessness impede growth and development,
discouraging foreign investment and domestic economic activ-
ity. Urban insecurity impacts productivity in several ways. In
many cities, employees resist working or leaving work after
dark when the streets are more insecure. Employers and
investors are less likely to invest in cities where their assets are
likely to be destroyed or stolen. This, in turn, limits the assets
and livelihood sources of the poor. Crime and the fear of crime
curtail urban investment. Both individual improvements in
standard of living – as minor as acquiring a radio or painting
a room – and entrepreneurial investments in buildings and
services are hindered by the likelihood of crime and violence.
In 2001, 61 per cent of surveyed firms in Kenya reported
experiencing criminal victimization. In such an environment,
businesses are forced to divert resources away from productive
147
LUIS FELIPE CABRALES BARAJAS
GATED COMMUNITIES ARE NOT THE SOLUTION TO URBAN INSECURITY*
Starting in the 1950s, many North American families, guided by a
pioneering spirit, decided to move their homes to fortified suburbs, an
attractive housing product promising a healthy environment, public
safety and lifestyle benefits. That urban process rapidly became
widespread during the 1970s, when it began to extend worldwide.
Because of the range of urban models and diversity of cultural settings
into which these urban spaces have been incorporated, there is no
standardized vocabulary to describe them. For example, in Spanish-
speaking countries they are called barrios cerrados, fraccionamientos
cerrados or urbanizaciones privadas; in Portuguese, they are known as
condomínios fechados and in English the terms walled communities and
enclosed neighbourhoods are used, although the most universally
recognized term is gated communities.
How can a distinction be made between legitimate
diversity and illegitimate inequality?
- Norbert Lechner
Implicit in the notion of a gated community is the decision to create
private urban spaces which are set apart from the rest of the city with
the aim of providing an escape from undesirable social disorders. They
have precedents in the 19th century, linked with the idea of garden
cities and the preference for moving to the suburbs, to which later were
added the rules of modern urban planning, aimed at dividing up the
urban space into single-use areas. Another type of urban planning that
aims to create protected surroundings is the design of cul-de-sacs or no-
through roads in areas that were previously open to traffic.
As a phenomenon that is part of the globalization process, the way
gated communities develop in different parts of the world tends to
reflect local economic conditions. In some countries, governments in
favour of keeping a tight hold on urban planning support social policies
based on redistribution, as is the case in some European countries,
which would explain why the trend is hardly noticeable – or even non-
existent – in those countries. The same cannot be said for Latin
America, where social divides coupled with the official permissiveness
of neo-liberal urban planning have been conducive to the development
of gated communities. Frequently, this has occurred in breach of urban
regulations that have proved incapable of controlling new processes
and has led to the privatization of streets and community areas that
were traditionally open to the public.
The image evoked by gated communities is rooted in ideological principles
and urban models that can be replicated: large or small housing estates
typically aimed at the middle and upper classes and usually surrounded by
a wall or fence. Those designed for the elite have sophisticated
electronically controlled security systems and police surveillance, club
houses, plenty of green spaces and sports grounds, and sometimes
include a golf course.
Luxury enclosed estates are usually based on architectural and urban
designs for low-density and low-rise housing, but they also occur in a
high-rise format. Swanky towers have been introduced in areas with
high environmental value in cities, as well as in suburbs such as Santa
Fé in Mexico City and coastal areas, such as Palm Islands in Dubai or
Miami in Florida.
These trends have been possible thanks to a combination of various
factors, such as growing insecurity and the deterioration of public
Dr. Luis Felipe Cabrales Barajas is a senior professor in the Department of Geography at the University of
Guadalajara, Mexico and is editor of the book, Latin America: Open Countries, Closed Cities.
areas, although reasons such as the desire for an exclusive address
and the guarantee of a high social status – either real or perceived –
also play a role.
All this has generated an ideological debate which has played into the
hands of real estate agents and private security companies. Set
against government inability to provide an effective guarantee of such
universal rights as public security or the provision of public areas,
these concepts are becoming commodities that are obviously
accessible only to a minority.
The real estate market both derives benefits from and promotes
paranoia about insecurity and environmental degradation, which
encourages the proliferation of enclosed neighbourhoods. The standard
approach is to mark out – both physically and symbolically – the
boundaries of the residential estate, adding value to it by providing
public areas and launching a good marketing campaign. This makes it
possible to push up the prices of the products – both houses and land –
and at the same time is conducive to speculative activities aimed at
converting the estates into lucrative capital havens. Notwithstanding
the high prices they pay, buyers are confident that their properties will
retain their economic value, provided that negative externalities are not
allowed to filter in.
The gated community shares many of the characteristics of
postmodernism: the privatization of urban services, the deregulation of
public utilities, individualistic practices, selective socialization, the
rejection of the best urban traditions and placing emphasis on the use of
private vehicles. If the value of urban development systems is assessed
from a very broad perspective, however, private – and, in particular, low-
density – urban planning runs counter to the fundamental principles of
sustainable development. Gated communities are major consumers of
land space and they conduct activities that constitute a wasteful use of
resources, including water and electricity. For example, in the
metropolitan area of Buenos Aires, enclosed urban estates occupy
approximately 30,000 hectares yet house only 1 per cent of the city’s
population; and in the case of Guadalajara in Mexico, enclosed
communities occupy 10 per cent of the city’s land space yet house only 2
per cent of the population.
From being an exceptional approach to housing until the 1970s, gated
communities have gradually become more widespread, and are now an
increasingly common model across the world and are a sign of changing
times for urban planning. The most conspicuous cases of urban planning
geared towards ensuring safety are possibly found in the southern
hemisphere, in polarized societies where the most affluent inhabitants
want to avoid contact with the rest of the city. The private cities of
Dainfern, Nordelta and AlphaVille, in South Africa, Argentina and Brazil,
respectively, are good examples of this.
Dainfern, which is located in Johannesburg, has an area of 320 hectares
and 1,208 houses with surface areas ranging from 450 to 1,600 square
metres. It has high-quality facilities on offer, including an on-site
university that carries the same name as the residential estate. The
security measures are operated through a network of 57 cameras and
the estate has a three-metre high electric perimeter fence
(www.dainfern.com). Nordelta – a sort of city-within-a-city, known in
Spanish as a “ciudad-pueblo” – located in the suburbs of Buenos Aires,
covers an area of 1,600 hectares and has nine private neighbourhoods,
three educational centres, a medical centre and artificial lakes
(www.nordelta.com). AlphaVille is an urban development located in São
Paulo that started in 1975 as an industrial zone and gradually became a
residential area. It covers a surface area of 500 hectares and
accommodates 50,000 residents as well as extensive commercial areas
and various facilities (www.alphaville.com.br).
Even if it is conceded that gated communities are a legitimate option
for the people who live there, it is important to recognize that they
pose new problems or heighten previously existing ones. Is it
beneficial to humankind that this model of settlements should
continue to spread? Are there sound arguments in favour of their
growing popularity? Do they really provide a solution to urban
problems or do they simply cover them up and masquerade them as
something else? However naïve or controversial they might be, the
possible answers to these questions will be useful for the purposes of
picturing future scenarios and directing new urban policy strategies
that foster social cohesion.
Given that issues such as the impact of gated communities on social
segregation, urban fragmentation, vehicular mobility and the
consumption of natural resources must be analysed, local governments
should discuss the appropriateness of these communities, reaffirming
the democratic principles to be applied in efforts to achieve cities that
are more inhabitable and a world that is less exclusive in the future.
Sources: Améndola, et al. 2000; Garay 2000; Glasze et al. 2005.
* Translated from Spanish
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Urban Crime Trends
The analysis of international crime trends is often
made difficult by the limited information available,
especially as regards time series. Most frequently,
statements about crime levels are not based on any
statistical evidence but are drawn exclusively from
media reports and local perceptions. Data is scat-
tered among different sources, is hardly comparable
(especially when referring to police statistics) and is
rarely available at the city level – particularly in
developing countries. Yet, large urban agglomerations
around the world share certain characteristics that
make comparison of crime trends across cities a
meaningful exercise, often more informative and use-
ful than those across countries.
Between 1989 and 2000, the International Crime
Victim Survey (ICVS) collected data on victimization
experiences of citizens from large cities and urban
areas in more than 70 countries around the world.
Crime situations surveyed by the ICVS were quite
general, but indicated an array of possible victimiza-
tion experiences that are more likely to occur in urban
contexts than in rural areas.
Data collected between 1992 and 2000 in 33 cities
and urban areas (populations of more than 100,000)
reveal an overall trend towards a decrease of victim-
ization experienced by citizens. The average percent-
age of respondents who experienced any crime in the
year preceding the survey went down from 32 per
cent to 29 per cent. A sharp decrease was observed
in 20 out of 33 cities and urban areas, while 4 cities
were stable and only 9 showed a marked increase.
However, in only 6 out of 20 cities and urban areas
was a major decline in actual victimization matched
by citizens feeling more safe. Citizens continued to
feel less safe in Buenos Aires, Bogotá and Warsaw,
and in the large cities of the United Kingdom, the
Netherlands and Switzerland, even though victimiza-
tion rates dropped. On average, more than half of the
citizens surveyed in the 33 cities and urban areas felt
either “very safe” or “fairly safe” (57 per cent). This
percentage remained unchanged between 1992 and
2000.
Declines in victimization rates were primarily attribut-
able to decreased frequency of house burglaries and
ordinary thefts. These crimes are among the most
common, and their reduction has a significant impact
on citizens’ quality of life. On average, burglary rates
Source:United Nations Office on Drugs and Crime (UNODC). Based on data from: United Nations Interregional Crime and Justice Institute, UNICRI (International Crime Victim
Survey 1992/96 and 2000); Home Office of the UK, Crime in England and Wales 2003/2004: Supplementary Volume 1: Homicide and Gun Crime; US Government, Bureau of Justice
Statistics, Sourcebook of Criminal Justice Statistics 2002.
150
went down from 4 to 3 per cent, while theft rates
decreased from 9 to 8 per cent. Indeed, house burgla-
ries decreased in most urban areas of developed
countries (the only exceptions were the cities in
Australia and Scotland), as well as Kiev, Tirana, Sofia,
Tallin, Warsaw, and Riga. A decrease in burglary
rates was also observed in cities in Asia, Africa and
Latin America (with the exceptions of Bogotá and
Johannesburg).
Trends in car theft, robbery and personal assault only
showed a slight decrease in average victimization
rates. Car theft decreased in approximately half of the
observed sites, including most capital cities in devel-
oping countries, while it increased in the capitals of
the new European Union Member States and in south-
east Europe. The average proportion of the surveyed
population that experienced car theft remained stable
over the study period, at 1.5 per cent.
Victimisation by robbery, which on average affected
approximately 2.5 per cent of the respondents,
increased in Buenos Aires, Johannesburg and Ulaan
Baatar, as well as in the capital cities of the new
European Union Member States, the United States,
Australia, the United Kingdom, France, Finland and
Sweden.
New York CELIA ANDERSSON
Finally, victimisation rates for personal assaults were,
on average, stable at 4.5 per cent. Personal assaults
decreased in cities in developing countries, North
America and South-Eastern Europe while increasing
in cities in Central and Western Europe and Australia.
Levels of personal assault and violence may be
closely related to homicide levels. Although the lat-
ter are not measured by the ICVS, homicide rates
recorded in police statistics largely match the
assault trends described above. Police data from a
group of 35 large cities, mostly in Europe and North
America, indicate a decrease in homicide rates in
the vast majority of cities (21) between 1998 and
2002. Homicide rates in 10 cities with populations
of more than one million people (Chicago, Dallas,
Houston, Las Vegas, Los Angeles, New York,
Philadelphia, Phoenix, San Antonio and San Diego)
continued to decrease between 2001 and 2002 from
an average of 13.2 to 12.5 per 100,000 people (with
homicide numbers falling in all 10 cities).
Nevertheless, the positive trend of the largest cities
was not matched by some smaller cities such as
Washington, Atlanta, Detroit, and Richmond, where
homicide rates were already high (above 30 per
100,000 people) and further increased between
2001 and 2002.
Homicide rates in the 8 cities with the lowest rates in
1998 (less than 2 per 100,000 people) increased the
most sharply to 2002, while in “high-risk” cities, the
rate remained relatively stable. This was the case, for
example, in Cape Town and Rio de Janeiro. It should
be noted, however, that homicide rates in these cities
were above 40 per 100,000 people, with some cities,
such as Medellin in Colombia and East London in
South Africa, showing homicide rates above 100 per
100,000 people. In such extreme cases, where homi-
cide represents the leading cause of death for juve-
niles, it is even more important to collect regular infor-
mation in order to monitor the impact of any initiative
aimed at crime reduction and crime prevention.
The available data series, although very limited, pro-
vides some reassuring signals on decreasing crime
levels in large cities, especially in developing coun-
tries. However, significant gaps still need to be filled,
since accurate and timely information is essential to
building proper strategies to combat urban crime and
insecurity.
uses, leading to reduced competitiveness and fewer invest-
ments – or even disinvestments. Crime therefore has a signifi- Endnotes
cant impact on economic development and investment.
While crime has many dimensions, strategies to tackle the 1 Although the attacks on New York and Washington are the most wide-
problem are often too narrow, focusing on tougher penalties ly known, several other cities, such as Nairobi and Dar es Salaam had
and law enforcement. A more effective way to control crime been targets of terrorism prior to 11 September 2001.
is to combine enforcement with prevention – developing 2 Cohen 2002.
positive strategies to tackle the underlying causes of crime. In 3 Department for International Development 2005.
Bogotá, Colombia, a multiple action plan for citizen educa- 4 Marcuse 2001.
5 Worldwatch Institute 2005. Estimates based primarily on United
tion focused on improving law, culture and morale; this con-
Nations-generated data.
tributed to a 30 per cent decline in homicide rates between
6 Ibid.
1995 and 1997. The programme targeted strategic areas such
7 Worldwatch Institute 2005.
as crime monitoring, police and judiciary action, education,
public services, youth programmes and restrictions on bear- 8 Appadurai 1996.
ing arms.28 9 Results from International Crime Victim Surveys conducted by UNICRI.
Urban violence, poverty and ineffective governance are inex- Comparative statistics available online: www.unicri.it.
tricably linked and mutually reinforcing. Unless urban vio- 10 WHO 2002.
lence – and its manifestation in exploitation of the poor – are 11 Results from International Crime Victim Surveys conducted by UNICRI.
addressed as part of poverty reduction and governance Comparative statistics available online: www.unicri.it.
improvements, programmes to improve the lives of urban 12 Vanderschueren 2000.
dwellers will have limited impact.29 Good governance both 13 Moser & Rodgers 2004.
supports and is supported by safe cities where inhabitants are 14 Narayan 2000.
free from fear. Poor governance increases the risk of insecurity. 15 United Nations 2005a.
Where safety is improved, interaction between the people and 16 Reported in the New York Times, 25 December 2005.
public institutions becomes possible, creating an enabling 17 Worldwatch Institute 2005.
environment for economic growth and participation. 18 Ibid.
19 Economic Commission for Africa 2002.
20 O’Higgins 2002.
■ Expanding the definition of human security 21 United Nations 2005a.
22 Vanderschueren & Vezina 2003.
The current concerns over crime and violence in cities 23 Moser 2005.
should not obscure the fact that in most developing country 24 Caldeira, 1996, writes, “the talk of crime is the principal discourse in
cities, the poor are contending with other forms of insecuri- São Paulo everyday life”. The perception that the institutions of order,
particularly the police, are also violent has magnified the fear and led
ty that threaten their lives and livelihoods. The security of
to the growth of private guard systems.
the poor, in particular, is affected by their health status,
25 Landman 2003.
which influences both their ability to work and their access
to health care. The HIV/AIDS pandemic has particular 26 Robertshaw, et al. 2001.
implications for urban security as it means loss of household 27 These trends have been supported by victim surveys in South Africa and
income, growth in the phenomenon of orphaned street chil- abroad.
dren, and disintegration of the family unit. Many urban poor 28 Kathuria & Oberai 2004.
families also face the constant threat of eviction. Insecurity is 29 Dwyer 2005.
exacerbated by insecure tenure with respect to both housing 30 Amartya Sen, quoted in CNN’s Principal Voices series in 2005:
and land. www.time.com/principalvoices.
As economist and Nobel Laureate Amartya Sen has noted,
“The demands of human security include a balanced view of
tragedies that are the result of terrible omissions as well as
commissions.”30 He defines an adequate concept of human
security in the contemporary world as one that includes the
following: a clear focus on human lives, as opposed to the
technocratic notion of national security in the military con-
text; an appreciation of the role of society and of social
arrangements in making human lives more secure; and a
fuller understanding of human rights, which must not only
include political freedoms, but also rights to food, medical
attention and basic education.
151
STEPHEN GRAHAM
Stephen Graham, Professor of Human Geography at Durham University, is the editor of Cities, War
THE URBANIZATION OF POLITICAL VIOLENCE and Terrorism: Towards an Urban Geopolitics, which was published by Blackwell in 2004.

Since the dawn of urban and military history, cities, warfare and organized rorist attacks. This is especially the case in rich, Northern cities. However,
political violence have always helped to constitute each other. As symbolic the almost infinite complexity and necessary openness of globalized cities
targets, urban centres have few equals. Contemporary cities are actually means that new solutions are unlikely to emerge beyond the highly
made up of almost infinite concentrations of sites, assets and spaces that restricted environments of airline systems.
can either be improvised as weapons projecting political violence them-
selves, or attacked as “soft targets” (entities that are not fully militarized or State military action against cities, on the other hand, is very much
equipped to fight back) by terrorists, insurgents and state militaries alike. defined by the technological capability of the military involved. In Africa,
Urbanites, moreover, are especially vulnerable to the disruptions caused by Latin America and Asia, a wide range of state anti-insurgency campaigns
political violence. This is because they rely on extensive concentrations of mobilize relatively unsophisticated technologies in their targeting of
technical and social infrastructure or capital to survive, feed themselves, urban insurgencies, their host populations, or increasingly militarized
access water and energy, avoid disease, remove wastes, and so on. gang and organized crime networks. Western and Israeli militaries, mean-
Disrupting these systems through political violence, deliberately or uninten- while, are being remodelled to adapt to the new demands of urban war-
tionally, leaves many urbanites with few alternatives. fare as part of their “war on terror”.

The post-Cold War period has seen a dramatic reduction in the number of As levels of urbanization around the planet continue to intensify, all
state-versus-state conflicts. Meanwhile, wars pitching state military or projections point to the deepening urbanization of organized political
paramilitary forces against non-state insurgent, terrorist, or organized violence in the future. As the sites, symbols and embedded assets of
crime groups have proliferated. This trend has been associated with a cities become both weapons and targets in increasingly mediated con-
dramatic urbanization of political violence around the world. Like other flict, a vital challenge for all concerned with the widest aspects of
facets of global social change, political violence is, in a sense, being human security is to resist the temptation to try and fortify cities
urbanized. More than ever, geopolitical concerns increasingly merge into against the putative risks in a narrow technical or architectural sense.
and irredeemably zero in on the very local sites and symbols of city life. Such a “fortress city” approach to “homeland security”, whilst lucra-
Not surprisingly, this change has been associated with a major change in tive to burgeoning military and security sectors, is a red herring
the balance of civilian rather than military casualties through political vio- because it is largely ineffective against determined attackers who can
Bank Underground Station, London PHILIPPE HAYS/STILL PICTURES

lence. Between 1989 and 1998, for example, approximately 4 million peo- simply select the next unprotected, soft target out of the millions of
ple were killed in violent conflicts around the world; an estimated 90 per options on offer in contemporary cities. Moreover, such an approach
cent of these were civilians – primarily women and children. also risks undermining the interchange, openness, flow and density
that sustain cities in the first place.
In an increasingly urbanized world, insurgent and guerrilla groups, rather
than seeking shelter within rural proletarian groups, are colonizing the The challenge, rather, it is to work at all scales of governance and conflict
world’s burgeoning urban spaces. At the same time – after centuries when mediation to try and ensure that the grievances, injustices, extreme ide-
cities were seen as sites to be either avoided or “rubbleized”, state military ologies and hatreds that fuel political violence against cities and urban-
doctrine, particularly in the West, now sees urban sites as the de facto ter- ites are, as far as possible, ameliorated. This must be done to the extent
rain for current and future struggle. Whilst they still occur, attempts at the that the murderous assaults on urban soft targets, by terrorists, insur-
complete annihilation of cities are now unusual, but the targeting of urban gents, and state militaries alike, are prevented or are rendered politically
soft targets as a means to coerce and win victory over a political enemy is or ideologically illegitimate.
now axiomatic to terrorists, insurgents and state militaries alike.
Such a challenge is daunting. This is especially so as urban research,
The methods and styles of this targeting could not be more varied. On the buses provided the intensely crowded and enclosed spaces necessary for geting. The worry here is that urban assaults merely radicalize the civilian policy and activism have tended to neglect the urbanization of political
one hand, non-state insurgent and terrorist groups increasingly exploit the suicide bombers’ actions. And in a widening range of suicide and car populations on the receiving end of violence, adding legitimacy to retaliato- violence thus far, leaving the subject to international relations special-
embedded assets of cities as weapons bringing instantaneous death, ter- bomb attacks in Indonesia, Iraq, Israel, Jordan, Kenya, Lebanon and ry attacks by terrorists. Moreover, in cities, combatants can target water, ists. But, in an increasingly urbanized world dominated by intensifying
ror and mediated violence. In the absence of sophisticated military hard- Morocco, among other countries, the unavoidably crowded spaces of sanitation, electricity, and food distribution systems through biological resource conflicts, global warming, proliferating refugee, water and
ware, the very bodies of volunteers are often mobilized to project violence urban everyday life are instantly being transformed into soft targets by agents and anti-infrastructure weapons such as missiles and bulldozers, food crises – sometimes precipitated by aggressive nation states and
directly against the sites and symbols of the modern city. terrorists. waging a “war on public health”. When this happens – as in Iraq after the transnational terrorist groups – the process fuelling the urbanization of
1991 Gulf War – far more civilians eventually die from preventable diseases political violence seems set to accelerate further. Through the rest of
Such projections of violence are becoming more spectacular and sophis- Whilst not reaching the levels of the total urban annihilation that character- than from the immediate effects of bombs and missiles. the 21st century, these challenges are likely to become even more criti-
ticated as the infrastructural and technological fabric of global cities ized 20th century warfare, targeting of cities raises concern because large cal. The time for a specifically urban treatment of geopolitics, which
intensifies in reach and complexity: airplanes became cruise missiles of numbers of innocent civilians are often killed as “collateral damage”. Cities The deepening sense of urban exposure and vulnerability, as transnation- concentrates on how local urban sites and infrastructure are enrolled
mass murder in New York and Washington; mobile phones were used to are unavoidably crowded and it is virtually impossible to distinguish al flows and networks erupt on city streets in violent acts of terror, has into global networks of political violence, is upon us.
trigger subway bombs in Madrid; and London’s underground trains and between insurgents and wider civilian populations, even with high-tech tar- provoked widespread search for technical or architectural solutions to ter-
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

4
Part Four

Policies and Practices

That Have Worked

Using slum estimates produced in previous parts of

the Report, this Part provides a policy assessment
of slum upgrading and prevention policies in more
than 100 countries across the world. An analysis of
the policy responses implemented by these
countries shows that the most successful ones
share similar attributes. Considering that policy
outcomes in reducing shelter deprivations have
been rather bleak, this Part highlights some
necessary bold actions needed to scale up
improvements today and prevent slums tomorrow.
It concludes by outlining key issues related to
development assistance provided by the
international community.
Road Construction, Bogotá, Colombia MARK EDWARDS/STILL PICTURES
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

4.1 Milestones in the Evolution of Human Settlements Policies FIGURE 4.1.1 SLUM POPULATION AND URBAN POPULATION GROWTH IN THE WORLD (1976-2006)

1976-2006 3,500

3,000

2,500

Slum and urban population (millions)

“Human settlements are linked so closely to existence adopt an integrated – as opposed to a sectoral – approach to
itself, represent such a concrete and widespread housing, infrastructure and basic services. The search for a
reality, are so complex and demanding, so laden with “harmonious integration of components” was an important 2,000
questions of rights and desires, with needs and aspi- outcome of Habitat I.
rations, so racked with injustices and deficiencies, 1,500
Urban population
that the subject cannot be approached with the
■ 1978: UN establishes focal point for human
leisurely detachment of the solitary theoretician.” settlements development 1,000
– Opening statement of Canadian Prime Minister Pierre
Trudeau at Habitat: United Nations Conference on Human
The importance accorded to human settlements at the
Settlements, Vancouver, 31 May 1976 500
Vancouver Conference was endorsed by the United Nations in Slum population
1978 when it created a special agency to serve as the focal
point for human settlements action and coordination within 0
1976 1988 1992 1996 2000 2001 2002 2006
the United Nations system. The United Nations Centre for
Habitat I Global Strategy UN Conference Habitat II Istanbul 3rd World
Human Settlements (UNCHS) was tasked with the responsi-
■ 1976: Habitat conference brings human Conference for Shelter on Environment
and Development
Conference plus 5 Urban Forum
Habitat 30 years
Millennium
settlements issues to the fore Declaration

World Summit
When the first United Nations Conference on Human on Sutainable
Settlements (Habitat I) took place in Vancouver, Canada, in Source: UN-HABITAT Global Urban Observatory 2005 and UN Population Division 2003. Development

1976, the world was largely agrarian, with two-thirds of world’s

three billion people residing in rural areas. Urbanization, as a
phenomenon, was beginning to be recognized, but more as a bility of ensuring that the shelter dimension was reflected in ed a fundamental role to the private sector in shelter delivery;
“problem” rather than as a positive force for economic, social UN programmes and projects. In its early years, the organiza- it was based on a sectoral approach that aimed to introduce
and cultural development. Cities were viewed as generators of tion (which later became the United Nations Human innovations in building technology, new construction methods
pollution, sites of unsustainable growth and a drain on nation- Settlements Programme or UN-HABITAT), worked with and affordable building materials. During this time, assistance
al resources. These negative sentiments were echoed by various other UN agencies to formulate shelter policies and imple- was focused mainly on the promotion of low-cost self-help
leaders attending Habitat I who blamed “uncontrolled urban- ment housing programmes in both rural and urban areas, par- housing, also known as the “bricks and mortar” approach. In
ization” for overcrowding, pollution and psychological tensions ticularly in least developed countries and those ravaged by theory, this approach offered many advantages: it allowed
and lamented “the disintegration of rural life and the disap- conflict or disaster. greater flexibility in building, and the possibility of communi-
pearance of farm lands through the spread of cities and their ty development and construction took place incrementally over
satellites”. time, giving beneficiaries the option of pacing the construction
The Vancouver Conference emphasized rural-urban dispari- ■ 1988: Global Shelter Strategy transforms role according to their household earnings. In practice, however,
ties and called on governments to improve the rural habitat by of the State from provider to enabler the approach had major shortcomings: it was entirely depend-
adopting policies that promoted a more equitable distribution ent on the supply of public land (or acquired private land), and
of the benefits of development between urban and rural areas. In the 1980s, scarce public funds and increasing urban pop- for many years remained project-oriented and limited to some
The Habitat Conference, Vancouver, 1976 UNITED NATIONS.

This led to a general “rural bias” among development agencies, ulations were putting a strain on governments in developing small-scale demonstration projects.
which focused their efforts on issues such as agricultural pro- countries, which soon came to realize that provision of public The GSS also came at a time when the idea of public provi-
ductivity and provision of basic services to rural areas. The housing was neither affordable nor sustainable in the long sion of services or the “welfare state” was losing legitimacy on
thrust of the recommendations in the Vancouver Declaration term. This realization led to the adoption in 1988 of the Global the global stage. Structural adjustment programmes (SAPs) that
was also more towards provision of public goods and housing, Strategy for Shelter (GSS) to the Year 2000, which advocated drastically reduced the role of the State in socio-economic
a policy that would change dramatically in the coming years. an “enabling strategy” that shifted the role of governments from development were being adopted by many countries in the
Nonetheless, the Vancouver Conference was the first milestone provider to “facilitator”. Governments were expected to remove developing world, which resulted in deregulation and privatiza-
in the “habitat” agenda; it recognized that human settlements obstacles and constraints that blocked people’s access to hous- tion of essential services. Unfortunately, public expenditure
– both rural and urban – were a new category of analysis and ing and land, such as inflexible housing finance systems and cuts in health, housing and education resulted in serious hous-
international policy intervention. Perhaps the most significant inappropriate planning regulations, while people were expected ing and basic services deficits that also increased levels of urban
aspect of this new habitat agenda was the recommendation to to build and finance their own housing. The GSS also accord- poverty in many parts of the world.

156 157
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
■ 1992: Sustainability emerges as a development
challenge
The 1980s also saw a shift towards democratization and
decentralization, which focused on strengthening the capacity
of local governments to improve urban management. By the
early 1990s, other non-state actors, such as civil society organ-
izations and the private sector, were clamouring for a greater say
in public affairs. These actors played an increasingly important
role in the development of international policies on human
rights and the environment.
Meanwhile, a 1987 report by the World Commission on
Environment and Development concluded that human activi-
ty was the leading cause of environmental degradation and pol-
lution. This led many governments to rethink their strategies
on environmental management. By the time the United
Nations Conference on the Environment and Development
(UNCED) took place in Rio de Janeiro in 1992, the concept
of “sustainability” (the idea that development should meet the
needs of present and future generations without harming the
environment) was gaining ground. However, while the envi-
ronmental movement focused its energies on addressing issues
such as biodiversity, global warming and desertification, it was-
n’t until the World Summit on Sustainable Development
(WSSD) in Johannesburg in 2002, that “sustainable urbaniza-
tion” emerged as a new multidimensional concept that covered
not only the impact of cities on the environment, but also their
potential to manage the urban environment in a way that ben-
efits urban residents both socially and economically. The move
towards sustainability spawned a variety of projects aimed at
improving access to basic services, such as water, sanitation and
waste disposal, among others, which were also the main areas
of focus of new joint initiatives, such as the Sustainable Cities
Programme and Localizing Agenda 21.
■ 1996: Globalization and urbanization shape
the Habitat Agenda
By the early 1990s, a series of United Nations conferences
focused the world’s attention on the challenges facing an
increasingly globalizing world. The second United Nations
Conference on Human Settlements (Habitat II), the last of this
series of conferences, took place in Istanbul, Turkey, in June
1996. By this time, 45 per cent of the world’s population was
already living in urban areas, with demographers predicting a
major shift in the world’s population from rural to urban in the
early years of the new millennium.
When world leaders met in Istanbul, globalization and
urbanization were becoming powerful forces of economic
growth, as well as exclusion. The rural idealism of the 1970s
had given way to the harsh urban reality of the 1990s where
cities were seen as engines of economic growth, innovation and
creativity, but also as sites of extreme poverty, exclusion and
environmental degradation. The world’s population had dou-
158
FIGURE 4.1.2 GDP PER CAPITA BY INCOME GROUPS 1975-2004
35,000
30,000
25,000
Dollars
20,000
15,000
10,000
5,000
0
1975 1985 1995 2004
High income World Middle income Low income
Source: World Bank, World Development Indicators database.
Note: GDP per capita based on purchasing power parity (PPP). Data is in current
international dollars. Income grouping based on World Bank's definition:
High income countries: those in which 2004 GNI per capita was $10,066 or more.
Middle income countries: those in which 2004 GNI per capita was between $825
and $10,065.
Low-income countries: those in which 2004 GNI per capita was $825 or less.
bled from 3 to 6 billion. Urban populations were growing at a
rate 2.5 times faster than rural populations, with the result that
almost half the world’s population was already living in urban
areas. The role of the State was diminishing as cities became
major centres of trade and finance, negotiating directly with
each other. Information and communication technologies were
changing the ways in which people, cities and countries were
communicating with each other. National borders were becom-
ing less significant and local authorities were becoming key
players in national and international negotiations. Civil society
organizations were also playing a more active role and demand-
ing to be heard at UN Conferences, including Habitat II.
Participation and partnerships were seen as the guiding prin-
ciples to “sustainable human settlements in an urbanizing
world” and “adequate shelter for all” – two of the main goals of
the Habitat Agenda and the Istanbul Declaration on Human
Settlements, which were adopted by the 171 governments at
the Habitat II Conference. Unlike the Vancouver Conference,
the Istanbul Conference affirmed the positive role of cities in a
globalizing world, and focused the world’s attention on urban-
ization as an emerging issue in the 21st century.
With the massive mandate set out in the Habitat Agenda,
UN-HABITAT struggled alone among United Nations agen-
cies to prevent and ameliorate problems stemming from rapid
urbanization and deteriorating living conditions. Although it
tried to address this shortcoming by forging partnerships with
other international agencies, such as the World Bank, the
impact of its programmes was not being felt among the most
vulnerable populations in cities – the urban poor. Part of the
problem was that while the Habitat Agenda offered the most
comprehensive guide to addressing these issues, it did not focus
the organization’s work in one particular area, which led to
duplication with other agencies. Also, without an agreed prior-
Burdur, Turkey DARCY VARNEY
itization and integration of recommendations with time-bound
targets, Member States of the United Nations were left with a
mission that encompassed all objectives, but no central objec-
tive. This made it difficult to monitor and review progress in
implementing the Habitat Agenda at the national, regional and
international levels.
■ 2000: World leaders set the “cities without
slums” target
In September 2000, the United Nations convened the
Millennium Summit where world leaders were expected to set
priorities and targets towards poverty reduction. This resulted
in the Millennium Declaration from which the Millennium
Development Goals and time-bound targets were derived. The
Millennium Declaration commits governments to addressing
key development issues and sets broad goals in order to eradi-
cate poverty by the year 2015. It encompasses key goals related
to poverty reduction, health, gender equality, education and
environmental sustainability.
The Millennium Declaration was adopted at a time when
almost half the world lived in urban areas, and almost one-third
of this urban population lived in slums. Governments were
increasingly recognizing that slums could not just be considered
an unfortunate consequence of urbanization, but needed to be
treated as a major development challenge through coordinated
policies and action at the global, national and city levels.
UN-HABITAT is mandated to monitor progress towards
achieving Millennium Development Goal 7, target 11 – “By
2020, to have achieved a significant improvement in the lives of
slum dwellers”. Known as the “Cities without Slums” target,1
this clear, time-bound target enables the organization and its
partners to re-orient their policies towards slum improvement
and prevention. Slum upgrading is now a very important area
of focus of UN-HABITAT’s, work, with increasing emphasis
being placed on policy and operational support to scaling up of
projects and programmes. Monitoring progress towards
achievement of this target at the national, regional and global actions are not put in place to alleviate, prevent or improve the
levels also comprises a significant part of UN-HABITAT’s housing conditions of the world’s urban poor. If no action is
work in this area. taken, the number of slum dwellers is expected to rise to 1.4
Despite the grand proclamations and commitments by gov- billion by 2020. It is hoped that 30 years after the “habitat”
ernments at the Millennium Summit, the reality on the agenda made its way into the development agenda, words will
ground suggests that poverty is rising in cities and that slums finally lead to the kind of concerted action needed to stem the
are emerging as a dominant type of settlement in many cities tide of rising poverty, exclusion and deteriorating living condi-
in the developing world. UN-HABITAT estimates indicate tions in cities.
that almost a billion people already live in slum-conditions
around the world and that slums are growing dramatically in
the world’s poorer cities, particularly in sub-Saharan Africa
and Asia.
Participants attending the 3rd World Urban Forum to cele- Endnotes
brate the 30th anniversary of the first Habitat Conference in
Vancouver in June 2006 will have to contend with this reality: 1 Millennium Declaration, paragraph 19.
the situation of slum dwellers is likely to worsen if policies and
159
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Vancouver: The World’s Most Liveable City Combines
Multiculturalism with Environmental Sustainability
“Vancouver is home to a multitude of cultures and languages from around the world. The City of Vancouver values this diversity, and
considers it a source of our strength, vitality and prosperity.” — City of Vancouver 2005
In October 2005, the Economist Intelligence Unit (EIU) voted
Vancouver, British Columbia, the world’s most liveable city.
Vancouver topped the EIU’s list of 127 cities around the world,
having earned the honour for its well-established infrastructure,
cultural and environmental richness, low crime rate, and easy
access to goods and services.
While many point to Vancouver for its environmental conscious-
ness and physical beauty, it is, in effect, the city’s cultural diversi-
ty and resources that work most effectively toward strengthen-
ing its environmental sustainability and liveability. Vancouver’s
population is a mosaic of cultures and ethnicities: 38 per cent of
the metropolitan region’s total population is foreign born, and
two-thirds of the adult population is directly connected to
immigration, being immigrants themselves or children of for-
eign-born parents. British Columbia’s largest city-region has
become a gateway to opportunity for people across the Asia-
Pacific region, an exemplar of Canada’s effort to build a cooper-
ative multicultural society and a leader in integrated planning
for liveability and sustainability.
Vancouver is a city of 550,000 residents set within a region – the
Greater Vancouver Regional District – that hosts a total popula-
tion of approximately 2 million. In 2001, Canada’s last census
year, half of the City of Vancouver’s residents were “visible
minorities”(non-Caucasian people who are not Aboriginal in ori-
gin), and 46 per cent were immigrants.The largest proportion of
the immigrant population comes from Asia – particularly China,
India, the Philippines, and South Korea.Vancouver is home to 14
per cent of the population of British Columbia but 24.5 per cent
of its total immigrants. Less than half the city’s population
speaks English as a first language.
Multiculturalism has served as a codified Canadian value and
policy framework since 1971, when the country expanded its
bicultural policy – which recognized the equality of the English
and French cultures in the development of the nation – to
include respect for and protection of the full spectrum of the
country’s citizens.The passage of the Canadian Multiculturalism
Act in 1988 further provided a comprehensive legislative struc-
ture for the country’s emerging multicultural identity. As one of
Canada’s most diverse urban centres, the City of Vancouver has
Sources: Mendes & Holden 2005; Leaf 2005; UN-HABITAT 2002a; City of Vancouver 2001; City of Vancouver 2004; Statistics Canada 2001; Economist Intelligence Unit 2005.
160
ELISABETH GATEAU
LOCALIZING THE MILLENNIUM DEVELOPMENT GOALS
initiated many strategies and partnerships designed to promote
social inclusion, increase civic engagement, recognize the many
cultural and linguistic groups in and around Vancouver, and
address the needs of Vancouver residents from a range of back-
grounds.
One of the major building blocks upon which multicultural pro-
grammes and services are built is the Civic Policy on
Multicultural Relations. Adopted in 1988 as a municipal exten-
Vancouver GOVERNMENT OF CANADA
sion of the national legislation, the policy recognizes ethnic, cul-
tural and racial diversity as a source of strength for the city and
resolves that all residents have the right to freedom from preju-
dice as well as the right to access civic services regardless of
background or language. A variety of institutional supports
within the city exist to help implement the policy.These include
a Social Planning Department that participates in the overall
planning of the city by reviewing city developments and provid-
ing advice on how to address specific cultural and social needs.
The Social Planning Department was instrumental in helping
diverse cultural communities and demographic groups partici-
pate in Vancouver’s component of the Greater Vancouver
regional planning process. The six-year participatory process
resulted in an award-winning Liveable Region Strategic Plan,
Greater Vancouver’s official regional growth strategy, adopted in
1996.
The City of Vancouver carried out its part of the regional process At the Millennium+5 Summit in New York, the United Nations General The new global realities of today’s fast-moving cities and diverse
between 1992 and 1995, when approximately 20,000 residents Assembly explicitly recognized the important role of local authorities towns and villages require that decisions be taken at the level of gov-
participated in the CityPlan initiative. As in the overall regional in contributing to the achievement of Millennium Development Goals. ernment closest to the citizens. Decentralization is essential for a
vision, the city’s residents expressed a strong commitment to The outcome document also opened up new possibilities for direct democratic system of governance and it is the key to basic service
connecting social and environmental principles for achieving dialogue between the General Assembly and the world organization provision for, by and with the community.
sustainability. They requested more efficient public transport, of local governments, and put local authorities first in the list of major
more and safer bicycle routes through the city and more pedes- actors that work with the United Nations. We must strengthen local governance, through enhanced citizen par-
trian-friendly streets. Residents also prioritized the following: ticipation and effective partnerships with all local stakeholders, if we
development of distinctive neighbourhoods featuring diverse As the United Nations Secretary-General stated when he met with a are to succeed in achieving the Millennium Development Goals. For
public spaces, affordable housing and access to services; a delegation of mayors: “A state which treats local authorities as part- this reason, 2007 will see the launch of a United Cities and Local
strong sense of community and increased public safety; and ners, and allows public tasks to be carried out by those closest to the Governments (UCLG) Global Report examining progress in decentral-
economic development that keeps jobs in the city. The city is citizens, will be stronger, not weaker. Weak cities will almost certain- ization and local democracy across the globe.
also investing millions of dollars to reduce homelessness and ly act as a brake on national development, whereas strong local
drug addiction within the Aboriginal population, which for democracy can be a key factor in enabling a country to thrive.”
many years was left out of Canada’s development agenda.
Despite these milestone achievements, there is still a great distance Elisabeth Gateau is the Secretary-General of United Cities and
to cover if we are to bridge the local versus global knowledge gap. Local Governments (UCLG), an international organization
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

■ Signs of commitment
4.2 Countries Taking Slums Seriously In some countries, political pronouncements have often acted
as a major driving force behind change, sending out explicit
signals for policy reform and unleashing innovation and initia-
tive in the delivery of shelter and services to the poor. This
commitment may be enshrined in the highest legal instru-

C
ommitment from the top1 obviously has some-
thing to do with why some countries have been
more successful than others in managing slum
growth. It is widely accepted that political will in
responding to the reality of slums is pivotal in
mobilizing commitment to help the urban poor to gain access to
adequate shelter, livelihoods and services. The experiences of 23
countries analysed by UN-HABITAT indicates that political
support for slum upgrading, slum prevention and urban poverty
reduction in general varies significantly over time among coun-
tries and within cities. Some national and local governments,
particularly during the last decade, chose to largely ignore their
burgeoning slums or confine actions to symbolic gestures, often
preferring to continue with practices of slum clearance and mass
slum upgrading efforts at national and State levels, the overrid-
ing concern seems to be with creating an urban environment
conducive to attracting foreign investment and selling India as
an attractive partner in the liberalized marketplace. While there
are progressive projects and policies being implemented in the
larger cities, there is a parallel and continuing trend towards
evictions without adequate or appropriate relocation. Other
countries, such as Ghana and Morocco, also show more modest
support for upgrading, having recently stepped up actions to
tackle slum growth. More moderate political support in these
countries over the last 15 years appears to have held back gov-
ernments’ performance in achieving the kind of turnaround in
slum numbers seen in the top-performing countries, where
political commitment was consistently stronger. These “moder-
ments, such as the South African constitution and the
Brazilian city statutes, both of which are based on the princi-
ple of equity. In other cases, statements by heads of state, such
as the Royal Letter in Morocco, or the public commitment of
the Cambodian Prime Minister, have set the benchmarks for
urban poverty reduction and slum prevention.2 Presidential
decrees have also helped to set in motion pro-poor pro-
grammes and budgets. In Mexico, a presidential executive
order launched the “Habitat Programme” in 2003 as part of
the president’s agenda to address poverty, health and education
in Mexico’s larger cities.
However, executive proclamations only make a difference
when they are backed by long-term strategies with realistic
national targets for slum improvement, adequate budgetary alloca-

Hammamet, Tunisia JOCHEN TACK/STILL PICTURES

evictions. By contrast, other governments have taken the respon-
sibility squarely on their shoulders: making commitments
backed by bold policy reforms; scaling up upgrading pro-
grammes; and adopting urban planning measures and equitable
economic policies to prevent future slum growth. However, most
governments seem to lie somewhere between action and inac-
tion, sometimes experimenting with new forms of planning and
financing more in line with the needs of the urban poor, trying
to push through much-needed reforms, but ultimately holding
back on political commitment to make a significant impact. This
part of the Report analyses how some governments have demon-
strated clear commitment and leadership in dealing with slums.
■ Where political commitment to slum
prevention and improvement has made a real
difference
Brazil, Egypt, Mexico, South Africa, Thailand and Tunisia
stand out as countries that demonstrated consistent political
commitment over the years to large-scale slum upgrading and
service provision for the urban poor. These countries are the
same ones that have been most successful in reducing or stabi-
lizing slum growth rates in the last 15 years, as revealed by the
scorecard results reported in Chapter 1.3. Egypt and Tunisia
recorded falling annual slum growth rates of 1.6 per cent and
5.4 per cent per year, respectively, while in Mexico and South
Africa slum growth fell to only 0.5 per cent and 0.2 per cent,
respectively.
Other countries, while often showing considerable political
determination, were ranked slightly lower according to the cri-
teria set to analyse leadership. In Indonesia and the Philippines,
the political upheavals of the last two decades have somewhat
undermined slum improvement efforts. In India, despite major
ate” performing countries experienced sustained slum growth
rates ranging from 1.4 per cent to 2 per cent per annum.
However, it would be unfair to say that all the countries that
are struggling to cope with high slum growth rates in this cate-
gory have shied away from committing to change. In sub-
Saharan Africa, Burkina Faso, Senegal and Tanzania have in
recent years shown promising signs of growing political support
for slum upgrading and prevention that include reforms in
policies governing land and housing. For example, Tanzania has
embarked on land reforms and has shown a more tolerant and
responsive attitude to its growing informal settlements. Yet
translating new-found support into significant improvements
on the ground will no doubt take time.
So there appears to be a pattern: the stronger the political com-
mitment by national and local governments to slum improve-
ment, the more significant the progress in reducing the growth
of slums. Several questions remain, however. Exactly how have
political leaders and top public servants shown – or not shown,
as the case may be – their commitment to slum upgrading and
preventive measures, and how does this make an impact on the
living standards of slum residents? And, perhaps the most critical
question is: what motivates them to take affirmative action?
While the focus of this chapter is on political willingness by
national governments to act decisively, it is recognized that polit-
ical leadership at the city level is also vitally important in driving
slum improvement programmes and policies. In some countries,
cities, under dynamic leaders, have gone ahead with innovative
citywide upgrading programmes despite a lack of political and
institutional support from the centre. But for the most part,
national governments have the power to make or break slum
policies and programmes – it is their political support, legisla-
tion, reforms and macro-economic policies that create the over-
all environment and framework through which programmes and
policies can operate locally, as well as on a countrywide scale.
tions, and a programme of policy, legal and regulatory reforms to
meet the targets. Morocco provides a good example of a coun-
try which, through its ambitious villes sans bidonvilles pro-
gramme for in-situ slum upgrading and greenfield develop-
ment, set the goal of becoming a “slum-free” country by 2010.
One hundred thousand units of social housing are to be deliv-
ered per year through significant budgetary allocations. A few
countries in Latin America, including Brazil, Chile and
Mexico, have made concerted efforts to develop long-term
action plans for slum upgrading and urban poverty reduction
by setting clear targets and establishing institutions to ensure
that these plans are implemented. In Brazil, following the
introduction of the “Slum Action Plan” by the Secretariat for
Housing and Urban Development, the President established a
housing fund of $1.6 billion for financing new housing con-
struction and upgrading favelas. The action plan is starting to
make an impact with the upgrading of 30 slums and the
improvement or construction of 31,000 housing units is
already underway.3 In Colombia, the National Council of
Economic and Social Policies (CONPES) has made a public
commitment to improve the living conditions of slum dwellers.
The national goal is to reduce the proportion of people living
in slums from 16 per cent to 4 per cent by 2020.
Consistency in political commitment is crucial in mobilizing
long-term support for slum upgrading. The Government of
Tunisia’s consistent support for upgrading over the past three
decades has all but eliminated the national backlog, which now
stands at some 24,000 units.4 In 1997, the Government of
Chile inaugurated the Chile Barrio Programme to achieve a
substantial improvement in the housing and employment of
families living in informal settlements. This carried the exten-
sive regularization process that started in the 1980s further. The
foundations for these initiatives were laid in the late 1970s
when Chile implemented a comprehensive national housing
programme. In recent years, the barrio programme has utilized
70 per cent of the budget of the Ministry of Housing and tar-
geted 30 per cent of the poorest population.5 Since 2003, the
Government of Thailand has been implementing programmes
for the construction of homes for one million low-income
households in partnership with commercial and public banks.
This commitment is a continuation of almost three decades of
attention to low-income housing, going back to 1977 when the
National Housing Authority (NHA) was given the responsibil-
ity for it and carried on as the umbrella body for slum upgrad-
ing. Such long-term support for low-cost shelter and slum
upgrading has undoubtedly contributed to Thailand’s extraor-
dinary success in improving living conditions in slums – the
slum growth rate has fallen by an average 18.8 per cent per year
since 1990.
Another indication of strong political commitment for slum
improvement is the inclusion of upgrading and urban poverty
reduction policies in the national development agenda. This has
led to prioritizing the urban sector and slum upgrading within
national budgets and macro-economic frameworks. As part of
Jordan’s National Strategy for Eliminating Poverty and
Unemployment, adopted in 1998, a major new upgrading pro-

162 163
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Costing the Slum Target*
In 2003, the UN Millennium Project began a costing First, a general methodology was developed Goals, and were also very similar to the cost-
process for the various targets that had been established that could be used for all kinds of interven- ing benchmark rules used by private sector
in the Millennium Declaration, and laid down principles tions and to test all sorts of options. With engineering firms in tendering for projects.
and a general methodology for costing individual goals. staff from UN-HABITAT and the Millennium
For this purpose, they proposed a modified form of Project, a very specific set of interventions The average global costs for these activities
Millennium Development Goal 7, target 11, namely, “By and standards was established in line with were found to be on average about $1800
2020, improve substantially the lives of at least 100 mil- industry standards and best practices in con- per person assisted, for both new sites and
lion slum dwellers while deterring new slum formation”. structing human settlements at modest cost. upgrading, with governments paying about
In practice, a costing was sought for two “technologies A number of recent surveys of upgrading and $1090 and the beneficiaries meeting the
of intervention”: construction in different parts of the world remaining costs. Overall, the total ground
were reviewed to establish the average cost costs to governments and donors of meeting
a) for 100 million people in existing settlements, a typi- of each type of upgrading activity, which the modified target 11 and assisting 800 mil-
cal slum upgrading package was to be provided in line ranged from streets, lighting and kerbing, lion people was about $830 billion over 17
with recent programmes; through basic community physical and social years. The construction programmes for the
infrastructure, to the purchase of land and the participating developing countries would be
b) for the 700 million potential new slum dwellers for construction of basic dwellings. The assump- similar in size (relatively speaking) to those
whom new construction on vacant land was to be tions used were in strict accordance with in which China and South Africa have typi-
undertaken, to a modest level of quality. those used for other Millennium Development cally engaged.
* Estimate based on work undertaken by Joe Flood in collaboration with the UN Millennium Project in 2004.
gramme was launched focused on improving living and envi- Afghanistan has yet to fully develop and implement its new
ronmental conditions, as well as promoting employment urban strategy. The National Urban Programme is the Third
opportunities and vocational training. As a consequence, a new Pillar of the Afghanistan Development Forum 2005 and is
upgrading programme was launched and resulted in 29 new intended to focus on upgrading the living conditions of house-
sites being upgraded, which helped to improve living condi- holds in under-serviced informal settlements. Understandably,
tions for some 327,000 residents.6 countries such as Afghanistan that are emerging from conflict
The experiences of these and other countries show that a gov- have been less able to mobilize the institutions and political sup-
ernment’s positive stand on slums can set off a chain reaction port required for slum upgrading, as the focus is usually on
of new strategies, policy reforms, laws, institutional develop- building the capacity of institutions and governance structures
ment, and scaling up of existing programmes which, over time, that were destroyed or failed to perform during the conflict. In
improves access to services, shelter and employment among the both Afghanistan and Somalia, for instance, one of the top pri-
urban poor. Of course, the impact of these interventions orities of the new governments is to build the capacities of insti-
depends on many other factors, including the capacity of local tutions and to rebuild destroyed infrastructure, rather than
and national governments, economic conditions, the willing- improve the lives of slum dwellers per se.
ness of donors, or simply the scale of the slum problem. But
one thing is for sure; not much moves without a committed,
energized national government. In other words, countries that ■ A test of political leadership: Recognizing the
have the political will also tend to elicit the capacity and right to the city
resources needed for slum upgrading and prevention.
At the other end of the spectrum, a number of countries have As the experiences of a number of countries show, govern-
shown weak political support for slum upgrading and urban ment commitment that stays strong enough and long enough
poverty reduction. In some cases, there may be declared politi- can help to reduce the numbers of slum dwellers. Other coun-
cal intention but there is limited evidence to suggest follow-up tries, where political leaders once looked the other way, are
support through clear policies, programmes and allocation of also getting more serious about slums. Much depends on the
public funds; implementation is also usually weak. Some coun- political juncture at which countries find themselves.
tries appear not to prioritize shelter and services for the urban Democratization and decentralization hold the promise of
poor at all. Afghanistan, Bangladesh, Ethiopia and Haiti are shifting the balance of power, giving the city’s poor a chance to
among the surveyed countries that fall into this overall category. bargain for a better life, but this remains a far distant goal in
In Haiti, for example, a national strategic plan on “Urban many countries. Meanwhile, the fact that centralized, and by
Development and Slum Upgrading” exists, and the Prime most measures “undemocratic”, governments have also man-
Minister has constituted a “National Commission on Social aged to make significant inroads into pro-poor shelter and
Integration” targeting slum dwellers, but not much progress services suggests that a top-down approach that is focused,
appears to have been made in implementation. Similarly, efficient and is backed by resources can also bring about posi-
164
tive change. Another force for change has been international
aid. Donors continue to exert influence on countries, bringing
pressure to bear on governments through aid for slum upgrad-
ing and urban poverty reduction – the issue is whether bilater-
al and multilateral development partners themselves are seri-
ous or not about stepping up aid for slums and how this might
further shape government attitudes.
City planning is about balancing and reconciling conflicting
interests. And, in the long run, cities are planned, built and man-
aged to reflect the values of those who hold power to make pub-
lic decisions. Slum dwellers have for too long been forgotten in
this equation. The present system – including the way official
local plans are made, land is allocated, and administrative rules
and regulations are set – is usually stacked against them and
instead favours better-off communities and bigger businesses. But
with a greater willingness to improve slums over the last few years,
governments, it could be said, are taking a more enlightened view
towards the urban poor and increasingly recognizing the “right to
the city” of squatters and slum dwellers in official circles.
But despite this sunnier mood, planners and politicians are
still caught between two worlds as they continually try to jug-
gle the interests of the poor and disadvantaged with those of
the urban middle class and elite. This tension is continually
being played out in the struggle for land, services and jobs –
and more often than not it is the poor who are left without.
City politics, in many places, continues to be dominated by
practices of political patronage – handing out favours and
services to certain communities in return for their political
loyalty – rather than on the basis of more objective criteria of
need and entitlement.
Even in countries ranked highly in this Report for their polit-
ical determination to provide for the urban poor, there is evi-
dence to suggest a complete about-face by the authorities at
particular points in time. More than anything this shows that
commitment to slum upgrading can be a fickle thing. A well-
meaning government that today shows firm support through
new pro-poor legislation, a major land titling push for the
urban poor, or a reinvigorated national upgrading programme,
could tomorrow authorize forced evictions of the very same
communities it sought to help.
Take the major cities of Rio de Janeiro, Jakarta and Mumbai.
The cities have shown, to some degree, political commitment
by adopting progressive, citywide policies and programmes –
helped significantly by reformed policy and enabling legislation
at the national level. In Rio de Janeiro, the Programa Favela-
Bairro resulted in an investment of more than $600 million
and has improved access to basic infrastructure, health and
education for nearly 500,000 people in the city.7 In Mumbai,
initiatives such as the Slum Redevelopment Scheme and Slum
Sanitation Project, built on partnerships among the govern-
ment, non-governmental organizations and community
groups, have also made a difference to the living conditions of
slum dwellers in the city. In Jakarta, the Kampung
Improvement Programme (KIP) has been hailed as Jakarta
Administration’s “best practice” response to slums in the city.8
And yet a closer look at these cities reveals that many of their
slum communities have recently gone through or are facing the
threat of eviction by the same authorities. In Rio, hundreds of
poor families are facing eviction as 14 shantytowns, the major-
ity in upper-class neighbourhoods, have recently been ear-
marked for removal by the city’s public prosecutor on the
grounds of preserving the environment, boosting tourism and
economic development, and diffusing urban violence.9 In
Jakarta, criticisms from some human rights groups have drawn
attention to the administration’s track record of evicting the
urban poor. According to some sources, between 2000 and
2005, the city evicted 63,000 people and a further 1.5 million
people are under threat in the wake of several new development
projects.10 Mumbai, above all, witnessed one of the most spec-
tacular shifts in policy as 90,000 shanties fell under bulldozers
in late 2004 and early 2005, all in the name of becoming a
“world-class” city.
Leaders in office today face a daunting task to respond to the
pressures around them from different interests. In committing
to change, many of these leaders are helping to kick-start a new
generation of policy reforms and large-scale programmes tar-
geting the urban poor. This has taken political courage because
it means convincing those who hold power to share their
wealth, land and other resources with those less powerful in
society.11 At times, the balancing act becomes a hard one to
maintain. In the face of mounting pressures to make their cities
more competitive or to respond to the demand for high- and
middle-income housing or commercial interests, authorities
can often slip back to policies and measures that once again
exclude the city’s poor – and, in the most extreme cases, revert
to some of the worst types of evictions and demolitions experi-
enced in their history.
Endnotes
1 This statement refers to political leadership at city level, but perhaps
more importantly at the level of central government.
2 Tebbal 2005.
3 UN Millennium Project 2005b.
4 Cities Alliance 2003.
5 Government of Chile 2004.
6 Hiasat 2005.
7 Inter-American Development Bank 2004.
8 Darrundono 2005.
9 Philips 2005.
10 Urban Poor Linkage (UPLINK) in Indonesia: www.uplink.or.id.
11 Biau 2005.
165
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7

4.3 Pro-poor Reforms on Slum Upgrading and Prevention Improving urban planning and monitoring in the city of Aleppo
In Syria, the planning process has traditionally been
guided by central planning authorities reinforcing
hierarchical procedures for reporting and approval
for projects. The country’s approach to urban man-
agement, which has consisted of comprehensive
master plans and regulations aimed at directing
decentralized system of monitoring urban space,
dividing the city into nine sectors, each with its
own monitoring unit and maintenance group. This
has allowed a closer relation between the munici-
pality and residents, in addition to improving the
city’s urban management scheme. In this regard,
and preserving its architectural heritage, con-
ceived of this task within a broader structure of
urban management, including land use regulation,
housing, technical infrastructure, traffic, and oth-
ers. Through an action areas approach, the project
engaged various local stakeholders such as non-

U
N-HABITAT’s policy analyses show that coun-
tries performing well in managing slum growth
have strategically targeted investments, legisla-
tion and pro-poor policy reforms in tackling
basic shelter deprivations: the absence of secure
tenure; overcrowded houses; poor durability of house construc-
tion; and the lack of safe drinking water and adequate sanita-
tion. Such reforms have gone a long way towards enabling cen-
tral government bodies, local authorities and urban poor com-
munities themselves to improve people’s access to land, housing
and basic services. In this Chapter, we examine some of the pol-
icy reforms that have made a positive impact on the lives of slum
dwellers in various regions. It should be noted, however, that
many of these policy reforms and programmes have yet to reach
an appropriate scale to deal with current deficits and future
needs in housing and basic service delivery.
Increasingly, a number of countries are offering pragmatically
designed tenure rights based on a spectrum of formal and infor-
mal legal arrangements, ranging from formal titling to custom-
ary rights of tenure. In parallel, these countries are trying to
ensure an adequate supply of well-located, affordable serviced
land that can increase the housing supply in the future and pre-
vent the growth of new slums. They are attempting to make the
land market work better by removing or reducing the legal and
administrative rules and regulations of both central and local
governments; this includes establishing a simpler land registra-
tion process, increased flexibility in approved building materials
and standards, and reduced minimum plot sizes and infrastruc-
ture standards. Through land-use planning and zoning, more
progressive local authorities are taking action to increase land
supply for future low-income housing and economic activities.
Better performing countries are also increasing investments in
water and sanitation and establishing pro-poor policies and
reforms in the sectors, allowing basic services to be provided at
rates affordable to the poor through appropriate design and
innovative structures of tariffs and subsidies.
Latin America and the Caribbean
Latin America and the Caribbean stands out as a region that has
adopted various pro-poor policies and instruments to integrate the
urban poor into the legal and social fabric of cities. Several coun-
tries in Latin America and the Caribbean have moved towards pro-
gressive national housing programmes and policies and land
reforms resulting in a marked improvement in the provision of
adequate low-cost housing with secure tenure. Governments across
the region are increasingly prioritizing land allocation to meet the
needs of low-income families and people living in informal settle-
ments with a long-term view that considers future growth and
slum prevention. Some of the best examples of national policies
and reforms come from Brazil, Chile and Colombia.
For example, Brazil has been a leader in establishing innovative
enabling instruments that have helped to improve the land devel-
opment and housing rights of the urban poor. The country’s
recently-adopted City Statute provides the legal and guiding
framework for municipalities to implement land management
and regularization processes. Several cities have already success-
fully used the provisions of the City Statute to regularize infor-
mal settlements and provide secure tenure to the urban poor.
They have sought to address land speculation by including spe-
cific measures for the compulsory use of non-built underutilized
or non-utilized land, and in case of non-compliance, they envis-
age progressive imposition of property taxes. In 1996, Brazil also
established an innovative planning and zoning instrument called
ZEIS (Zone of Special Social Interest). A ZEIS is local authority
demarcated area which allows the application of flexible stan-
dards to promote upgrading.
Profavela is another piece of pro-poor legislation that is help-
ing low-income communities in Brazil to gain access to secure
tenure. The Profavela federal law has been adopted at the local
level by the city of Belo Horizonte and is currently enabling slum
dwellers to negotiate with public authorities and service
providers to establish a plan for the improvement and regulariza-
tion of their settlements. Belo Horizonte has 177 slums and 63
public housing projects, totalling 240 low-income neighbour-
hoods housing approximately 500,000 people. The legislation
enables the local authority to suspend and relax, on a temporary
basis, relevant laws, by-laws and standards to facilitate improve-
ments and land tenure regularization. Once the plan for
improvement and regularization is approved, it becomes a legal
instrument that further enables public intervention, and empow-
ers slum dwellers to invest in improvements and to comply with
agreed codes and standards, and ultimately, to gain legal recogni-
tion and title deeds.1
However, despite a long history and commitment to improv-
ing the lives of slum dwellers, Brazil has been unsuccessful in
improving the lives of the poorest of the urban poor; inequality
and chronic poverty are on the rise, and perceptions about favela
(slum) dwellers have not changed. One recent study in Rio de
Janeiro found that there is still a pervasive stigma against favela
dwellers in the job market; in fact, living in a favela seems to be
bigger barrier to gaining employment than being dark skinned or
female.2 Clearly, there is a need to change people’s perceptions
about slum dwellers and institute reforms that go beyond slum
growth and organizing service delivery at a macro- the municipality has set up its own “Local Urban governmental organizations, citizen groups, and
scale, does not allow flexibility for municipalities to Observatory” as a focal point for developing an other state agencies and was hence able to extend
decide on a wide range of functions. Most munici- information network to support planning deci- its partnership network even broader. The project
palities also lack the capacity to deal with the con- sions. The observatory is helping the city to collect led to substantial improvements in the living con-
siderable overloads in their daily activities. data in order to create a more accurate profile of ditions of the old city’s inhabitants and at reducing
the current urban situation, to identify major chal- neighbourhood degradation. This plan provided an
However, recently a number of municipalities have lenges and areas of intervention, and track important departure from earlier centralized mas-
been able to overcome some of these challenges. progress towards sustainable urban development. ter planning experiences towards a more flexible
The renowned old city of Aleppo, famous for its development plan in priority action areas. The
history and monuments, is a good case in point. The city of Aleppo has also implemented an municipality has also been able to involve several
The municipality has been able to introduce a new upgrading scheme for its old city core that has stakeholders and generate innovative partnership
set of codes and regulations in order to improve its given it widespread visibility and recognition for with local and international actors that allowed it
built environment and the living conditions of its its planning practices. The new planning scheme, to compensate for lack of know-how as well as the
residents. The municipality has also initiated a which aims to upgrade the historic core of the city scarcity of its resources.
Source: ESCWA 2001, UN-HABITAT 2004.
Economic and Social Commission for Western Asia, Sustainable Urban Development: A regional perspective on good urban governance, United Nations, 2001
World Urban Forum 2004, Barcelona, Presentation on Aleppo Local Urban Observatory, Mayor Mann Chibli.
upgrading and regularization in order to safeguard poor people’s subsidies in an attempt to reach the poorest 20 per cent of the
livelihoods. population and to meet the rising costs of social housing; this
The Government of Brazil is also implementing progressive policy has been credited with reducing poverty levels in urban
reforms in the provision of water and sanitation. The National areas. In parallel, the country has also instituted a national land
Sanitation Policy emphasizes environmental sanitation as a tenure and titling program called “Póngale titulo a sus sueños”.
social right and the adoption of new regulation and inspection
guidelines that establishes the rights and obligations of all Asia
providers and users of services. The government has also sig-
nificantly raised investments in water supply and sanitation The performance of Asian countries in carrying forward
systems.3 It is estimated that these investments are likely to major pro-poor reforms and programmes in land and housing
benefit 9 million families across the country. Furthermore, the provision has also been generally good. In India, national policy
government is stepping up its actions to improve the capacity guidelines on housing are being finalized, drawing on the Urban
of sanitation operators.4 Land Ceilings and Regularization Act from the 1970s, which
In Colombia, the new constitution in 1991 transformed the allowed municipalities to set aside land for the shelter needs of
way land was utilized in urban areas. The Law on Spatial the urban poor. Individual states and cities have undertaken
Planning that followed is based on the overarching principle of innovations such as the use of transferable development rights to
protecting the public over individual interest – this has led to a free up land for low-income housing. In 1996, India’s Slum
more rational use of land, greater equity in the provision of basic Rehabilitation Act allowed state authorities to offer land devel-
infrastructure and services, the protection of the environment opment rights to slum and pavement dwellers.
and the preservation of cultural heritage. In cities such as Bogotá India is also implementing reforms that go beyond the hous-
and Medellin, it has enabled the adoption of innovative practices ing sector, but which have the potential to significantly improve
in the integration of low-income settlements, as well as a more the lives of slum dwellers. India’s five-year development plan
efficient network of roads, infrastructure and transport. A key aims to promote universal coverage of water supply and sanita-
success factor has been the instrumental role of the Ministry of tion. In pursuit of this goal, several central government-spon-
Economic Development in nationwide campaigns to familiarize sored schemes and programmes have been implemented. Under
the public with the law. this programme, by the end of the first quarter of 2005, a total
Other countries have experimented with large-scale govern- of 5 million sanitation units were constructed. Similarly, the
ment subsidies to potential home-buyers and developers. For National Slum Development Programme (NSDP) looks specif-
example, Chile reformed its housing policy in 2001 to increase ically into upgrading of urban slums by combining physical

166 167
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
infrastructure with social services, including water supply,
community latrines, storm water drainage, community bath-
rooms, sewers and other amenities. The government has also
recently launched a new programme – the Jawaharlal Nehru
National Urban Renewal Mission (JNNURM) – that aims to
bring about mandatory reforms both at State and municipal
levels to improve basic service provision and secure tenure in
urban poor neighbourhoods. The programme, the single
largest national government initiative in the urban sector, was
launched in December 2005 and is to be implemented over a
period of 7 years. The programme’s special components
include water supply and sanitation, sewerage and solid waste
management, construction, and improvement of drains and
storm water drainage. However, because of massive backlogs in
housing and basic services, and because of high levels of
urbanization, these important programmes may prove insuffi-
cient in the face of the huge challenges that they need to
address, particularly with regard to annual slum growth rates
that are estimated to be 1.72 per cent per annum.
In Sri Lanka, the provision of housing has been considered
a major priority by successive governments since independ-
ence in 1948. In 1977, the government ventured into signifi-
cant and ambitious attempts to increase the housing stock in
the country and increase home ownership. A distinct and sep-
arate authority, the National Housing Development
Authority, was established to implement and promote mass
housing programmes such as the One Hundred Thousand
Housing Programme and the One Million Housing
Programme, and provide long-term subsidized loans for new
developments and upgrading activities. In 1994, it focused its
attention on high density housing in urban areas. In view of
increasing land scarcity, the government decided to stop slum
upgrading and initiate a programme to relocate slum and
shanty dwellers in apartments built by the government. The
Indian Ocean tsunami of December 2004 had a deep impact
on housing in the country’s coastal towns, but the govern-
ment’s commitment to rebuild houses and communities has
remained unaltered, which bodes well for the island nation’s
prospects for recovery.
Cities in Thailand have adopted different kinds of innova-
tive land-use mechanisms. Local context-specific solutions are
designed with close guidance from government, community
groups and NGOs. In the 1980s the ministry of finance reor-
ganized the Government Housing Bank, which is now a lead-
ing institution in Thailand’s housing finance system. It has
improved housing affordability in the country and enabled
large sections of the population to buy houses. In the past
seven years, the economy of Thailand has grown at a rate of
more than 8 per cent per annum. Since public sector housing
has become very common, the private sector has had led to
lower the costs of housing, which is making housing affordable
to the majority. The private housing market has also developed
its operations effectively and offers the lowest lending rates in
the market, challenging other financial institutions to lower
their interest rates in order to compete for business.
168
The recording and registration of urban land could be a first
step in the process, but there is also considerable potential for
developing systems that create an interface between informal
and formal systems. In some countries, this interface is occur-
ring at an incremental level, but still not on a scale that can
address the roots of the problem, which are structural, legal and
economic. Both Ghana and Tanzania are embarking on the
process on a pilot basis. In Ghana, following a long period of
military rule and a slow shift back to civilian rule, the housing
markets have become severely crippled. A substantial portion of
all land and housing in the country is informal. Land titling
and registration are major bottlenecks in slum upgrading and
work on this is beginning in two of the bigger cities, Accra and
Tema. In Tanzania, the 1995 land policy recommends registra-
tion and recording of all urban land and states that “existing
Bangkok, Thailand ©PHIL DATE. IMAGE FROM BIGSTOCKPHOTO.COM
squatter areas …will be upgraded”. The 1999 Land Act initiat-
ed the process of registering all properties in unplanned areas.
So far, 3,000 titles have been issued to people who hold prop-
erties and land in the slum areas of Dar es Salaam. The main
objective of the project is to enhance land tenure by issuing res-
idential licences to slum dwellers. The Tanzanian government
also embarked on another project in 2002 to allocate 20,000
serviced plots to residents who would have otherwise built
housing in unplanned areas.5
Reforms in Burkina Faso and Senegal, on the other hand,
appear to have a longer history and run deeper. Measures that
Burkina Faso took from the early 1980s to address informal set-
tlements were very much a result of a central government deci-
sion to undertake large-scale physical and tenure upgrading of
all irregular settlements in the capital city. Although this was
done through the nullification of individual titles (an exercise
that was reversed later), a culture of improvement and accept-
ance of informal dwellings did set in. After 1991, new reviews
Sub-Saharan Africa of land legislation brought back the land titles abolished by the
upgrading process of 1984 –1990. De facto security of tenure
Africa’s policy experiences with improving tenure security and was enforced and further reforms have oriented land manage-
access to land have been mixed, and fall way short of the ment towards a market approach. In 1997, the National
progress made in Latin America and Asia. Post-independence Company for Urban Land was set up for the provision and sale
land reform was postponed in countries undergoing political of serviced land. Despite these reforms, both annual slum
upheaval, military rule and experiments with different forms of growth rates and slum prevalence are still very high (4 per cent
government. The emphasis was on rural development, and it is and 76 per cent, respectively). In 1986, the Government of
only now that governments in the region are beginning to Senegal attempted to implement a tenure regularization pro-
address land and housing for the poor in urban areas. Many of gramme at the national level. Physical upgrading and tenure
the region’s problems stem from the fact that many countries regularization were carried out in parallel. In effect, this estab-
have inherited tenure and administration systems that are not lished a tradition of tolerance for informal settlements and led
appropriate or relevant to the needs of modern African cities. to low rates of eviction.
Formal systems of land registration and administration have In 1994, after the first open democratic elections, the
been unable to cope with rapid urban growth, with the result Government of South Africa promised to build one million
that between 50 per cent and 70 per cent of all urban land in houses a year. In order to reduce the housing backlog, the gov-
the region is delivered through informal systems.4 Governance ernment established a number of social housing institutions
failures, lack of economic development and industrialization in accompanied by a people-driven housing process, but housing
some countries, rising urban poverty levels and high proportions provision has not lived up to the promise. While the supply of
of slum populations in cities have all compounded problems housing increased between 1994 and 1998, it actually
associated with securing tenure and gaining access to land and decreased between 1999 and 2004. Although considerable
housing in urban areas. work is being done to improve security of tenure, scarcity of
urban land has forced the government to place urban poor
households on the outskirts or outside the main commercial
centres, which impacts their ability to earn a living. In many
areas, racially-segregated distribution of land in the apartheid
era still dictates urban landholding patterns. While there is
impressive expansion of housing stock, urbanization is also
escalating, creating new situations of unmet demand. The need
to build skills at the local and central levels to deal effectively
with informal settlements upgrading and tenure provision
remains urgent. South Africa has found that despite sincere
intentions to fast-track social housing and upgrading, the
absence of adequately qualified technical personnel creates a
major constraint.
Reforms in other sectors are beginning to make an impact.
South Africa stands out as a country that has made universal
access to water and sanitation a high priority. In order to redress
the imbalance in access to water and sanitation brought about
by past apartheid laws, the government has been thoroughly
reviewing its national policies and legal frameworks so as to
ensure that all new legislation conforms to the principles of
equity, fairness and sustainability. A new water law provides
every household with 200 litres of free water per day, which has
substantially increased coverage in urban areas.
In other parts of the continent, there are growing signs of
more progressive water and sanitation policies by governments.
For instance, the three East African countries of Uganda,
Tanzania and Kenya are undergoing radical water sector
reforms. The key objectives of the reforms are to promote good
governance and improve the performance of the sector. The
reform process has followed a similar route in the three coun-
tries: separation of water services and water resource manage-
ment activities; separation of asset-holding and development,
service provision and regulatory activities; decentralization and
devolution of responsibilities to the lowest practical level (sub-
sidiarity); greater transparency through increased civic engage-
ment; reduction of political interference at all levels through
the introduction of commercialized operations, and public-pri-
vate partnerships, where appropriate.
Endnotes
1 UN-HABITAT Best Practices Database: www.bestpractices.org.
2 Perlman 2005.
3 Government of Brazil 2005.
4 Rakodi 2005.
5 Based on unpublished briefs prepared by the Tanzanian Ministry of
Lands and Human Settlements.
169
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
4.4 Governing from the Bottom, Governing from the Top,
Connecting the Two
■ Local governance takes centre stage
T
he concepts of good urban governance and the
“inclusive city” have gained considerable cur-
rency in recent years. In particular, UN-
HABITAT has been a major campaigner for
inclusive, participatory decision-making in
cities and devolution of power from central to local govern-
ments – two of the cornerstones of good urban governance. 1
In practice, improvements in urban and local governance
have taken different forms in different countries and regions
of the world. The major transformation in local governance
over the last decade or so has been through the process of
decentralization. The conceptual basis for decentralization is
provided by the notion of subsidiarity, which implies devolu-
tion of responsibilities to the “lowest appropriate level”. The
expanded responsibilities of local governments as described
above are a result of the devolution of both power and
responsibilities from national to local governments. They are
primarily a consequence of administrative decentralization,
wherein decision-making authority and responsibilities are
transferred to sub-national governments. Administrative
decentralization is often preceded or accompanied by politi-
cal decentralization. Financial decentralization, on the other
hand, is the most complex step in the decentralization
process. It is often the most contentious as well, as national
and local governments struggle to retain and wrest control of
local budgets. Decentralization exists in its most advanced
form when elected local governments are empowered and
capable of setting development priorities, making major
development and expenditure decisions, and determining
and collecting local revenue. The other critical trend in local
governance in recent years, as a spin-off to decentralization,
is the growing trend towards direct, broad-based participa-
tion of communities in decision-making as way of improving
responsiveness of local policies and initiatives to citizens’ pri-
orities and needs.
While many developing countries are now preoccupied
with carrying forward reforms in decentralization and trying
to give communities a greater voice in local planning, it is
worth considering for a moment the state of play – who’s
actually doing well in making important strides towards good
local governance? And is any of this making a difference in the
lives of slum dwellers?
170
■ Empowering cities: A look at country
performance in building local governance
Where local governance is becoming a reality
By most criteria, several Latin American countries, including
Brazil, Colombia and Mexico, take the lead in the area of
improved local governance. The process of “re-democratization”
in the late 1980s and early 1990s in Latin America resulted in the
adoption of progressive policies aimed at reinforcing local govern-
ment and promoting inclusion by allowing grassroots movements
to take part in decisions at the local level. Numerous initiatives,
such as participatory budgeting, participatory planning, popular
movements for access to land and housing, and empowering
women with a greater voice and choice in local governance have
emerged from this region. Brazil was among the first countries to
introduce “participatory budgeting”, an innovative mechanism in
representative democracy that allows community-led city councils
to decide on health, education and other policies and on the allo-
cation of municipality budgets. In Belo Horizonte, for instance,
up to half the local resources for investment were allocated using
this method in 1999. In other cities, participatory budgeting has
resulted in better and more social services. Since the city of Porto
Alegre adopted participatory budgeting in 1989, for instance, the
number of public schools has risen from 29 to 84 and the propor-
tion of the population with access to the municipal sewer network
has grown from 46 per cent to 84 per cent.
Participatory budgeting has been praised, both nationally
and internationally, as a shining example of good governance.
By 2002, over 140 of 5,571 municipalities in the country had
adopted participatory budgeting as a policy. Although each city
adopts different formats to define investment criteria and to
select community representatives (who are generally from low-
income districts), the process has resulted in more active partic-
ipation of civil society in municipal decision-making.2 The
revised 1988 Constitution also decentralized resources to the
sub-national and local government levels, making it possible
for local governments to institute various pro-poor policies
aimed at integrating low-income communities into the fabric
of urban society. Through initiatives like the City Statutes and
participatory budgeting, Brazil has made major advances in
developing a participatory and sustainable way of city planning
and management, and has empowered the country’s 5,000 or
so municipalities.
Asia, as a whole, is not far behind Latin America in efforts
to improve local governance. In Southern Asian countries,
such as India, Nepal and Sri Lanka, municipal authorities are
semi-autonomous bodies with substantial independence from
central government, and have significant responsibilities for
the provision of public services.3 South-Eastern Asian coun-
tries are also starting to make progress on decentralization.
After shifting from military rule to a democratic system,
Indonesia began a decentralization process in 2001 in an effort
to give more political and financial clout to local governments.
In the Philippines, the adoption of the Local Government
Code in 1991 devolved to municipal authorities the responsi-
bility for basic services such as health, primary education, pub-
lic works and housing, and helped increase the financial
resources of local government.
Asia is also home to a strong civil society, focusing in partic-
ular on the rights of the poor and slum dwellers, but civil soci-
ety action has had a mixed record with regard to its impact on
the ground. In India, while a vibrant, organized civil society
liaises with government on innovative shelter solutions and
service provision to the poor, it has not been able to halt an
ongoing trend in some large cities of evictions without ade-
quate or appropriate relocation; more often than not, partisan
politics or commercial interests determine whether or not an
eviction is to take place. In other countries, community-level
participation has made a real change in people’s lives. The
strength of Sri Lanka’s slum upgrading approach, for instance,
lies in the tradition of keen community participation. For
example, programmes and initiatives that began as far back as
the 1970s, such as the Urban Settlements Improvement
Programme and the Urban Basic Services Programme, created
local community level institutions that continue to exist today.
But there are as yet inadequate channels for community-based
processes to link to government decision-making.
In Eastern Asia, Thailand provides a shining example of par-
ticipatory governance that has resulted in successful slum
upgrading efforts. In 1992, the central government initiated
the largest community-driven programme for assistance to the
urban poor in any developing country through its Urban
Community Development Office (UCDO). UCDO support-
ed community organizations with loans, small grants and tech-
nical support and encouraged them to form networks to nego-
tiate collectively with the city and provincial authorities. In
2000, UCDO merged with a Rural Development Fund to
form the Community Development Institute or CODI and it
continues to be a parastatal. CODI facilitates active dialogue
among communities in informal settlements and municipali-
ties, non-governmental organizations, and the private sector.
By promoting such partnerships at the city level between
municipalities, the agency helps to make sure that slum
upgrading projects are well designed and, as far as possible,
sustainable.
In sub-Saharan Africa, very few countries have attained a sig-
nificant degree of devolution. Most decentralization initiatives
are relatively recent, and many are poorly implemented due to
resource constraints and weak institutional capacity. In
Cameroon and Senegal, laws on decentralization were passed in
1996; Tanzania followed suit in 1999 and South Africa in
2000.4 However, some of the other African countries that have
attempted to devolve power away from the centre have ended
up giving significant authority to districts and regions, not to
local governments. In some countries, including Burkina Faso
and Senegal, decentralization policies have been developed but
implementation has been slow. For example, Burkina Faso
adopted a new constitution in 1991 that devolved administra-
tive powers and the authority to tax to the local level. In 1995,
a new decentralization law was passed that gave municipalities
a large array of responsibilities, including land management.
But in fact, this promise did not materialize as the central gov-
ernment has shown to be reluctant to hand over land manage-
ment responsibilities to municipalities. In countries where
there is a degree of decentralization, such as South Africa and
Tanzania, municipalities have often been stopped in their
tracks by a lack of funds, inadequate technical capacity, insuffi-
cient administrative resources and ambiguous regulatory guide-
lines on how to implement legal frameworks at decentralized
levels. In fact, wherever there has been progress on pro-poor
reform, it can be attributed to clear direction and support from
national government institutions.
■ Decentralized versus centralized governance:
What works best in improving slums?
Where bottom-up local governance works
Are countries that are getting better at governing their cities
from the bottom up also improving the lives of slum dwellers?
In Latin America, it appears that those countries where decen-
tralization and people’s participation is strongest, such as
Brazil, Colombia and Mexico, have performed well in stabiliz-
ing slum growth rates since 1990. Brazil and Mexico recorded
0.3 per cent and 0.5 per cent annual slum growth rates, respec-
tively, while Colombia registered a slightly higher growth rate
of 1.1 per cent. On the other hand, Asian countries appear to
have struggled somewhat more than their Latin American
counterparts in translating policies into significant improve-
ments in the lives of slum dwellers, despite implementing wide
reforms in decentralization. For instance, India, Nepal and the
Philippines registered slum growth rates of 1.7 per cent, 4.8
per cent and 1.9 per cent per year, respectively, since 1990.
Thailand, however, is one major exception in the region where
the number of slum dwellers has fallen by a remarkable 18.8
per cent per year and where, as shown earlier, there is strong
track record in community-driven upgrading with extensive
government support. Sri Lanka, too, has performed very well,
recording a decline of 3.7 per cent in the annual slum growth
rate since the start of the 1990s.
Sub-Saharan Africa lags behind Latin America and Asia in
efforts to improve local governance at the local level and it is
171
 TONY HILL
CIVIL SOCIETY AND THE URBAN AGENDA
Since the fall of the Berlin Wall in 1989 and the subsequent expansion of
more democratic forms of governance around the world, civil society, non-
governmental organizations (NGOs) and citizen’s groups of all kinds have
emerged in great number everywhere and have shown themselves to be a
vital force in tackling some of the world’s most pressing problems. Whether
it is a mass lobby for a better deal on aid, trade and debt for developing
countries, or the provision of services and material and moral support for the
poorest communities and people, or self-organized citizen groups demand-
ing their basic human rights, civil society has emerged as a key driver of pro-
gressive social, economic and political change in all regions of the world. As
UN Secretary-General, Kofi Annan, observed in 2004, “The partnership
between the UN and civil society is…not an option; it is a necessity.”
There is a growing recognition that the battle to achieve the Millennium
Development Goals, to eradicate poverty, to achieve gender equity and
human rights for all, and to move towards environmentally sound patterns
of production and consumption, will increasingly take place in the world’s
cities. Cities are prolific users of natural resources and generators of
waste, pollution and the greenhouse gases that cause climate change; and
with one in every three urban dwellers living in a slum, cities concentrate
and manifest extensive poverty and exclusion in some of its most shocking
forms. With demographers projecting that 60 per cent of the world’s popu-
lation will live in cities by 2030, up from around 30 per cent in the 1950s
and 50 per cent today, it is not hard to imagine the momentous challenges
that have to be faced in securing clean water supplies, waste and pollution
management, decent housing, employment, urban transport and so on, all
within a framework of law and respect for citizens’ human rights.
In response to these daunting challenges, new directions in urban gover-
nance, policymaking and action are beginning to emerge, based upon
greater recognition of the legitimate claims of different stakeholders
involved in urban issues and problems. This movement is bringing together
central governments, local authorities and municipalities, and civil society
organizations and groups in joint efforts to address the most pressing prob-
lems. This is manifest in the tremendous upsurge of different forms of inter-
national, regional, national and local alliances, coalitions and partnerships
focused on city issues around the world over the past 15 years. At the same
time, there has been growing understanding and acceptance that civil soci-
ety advocacy work on urban issues is a legitimate part of good governance
and democratic politics, and can lead to more just, effective and efficient
outcomes. Of course, this is not the case everywhere, and even where
these new forms of participatory politics are taking root, these are still
early days with much more work to be done if the dynamics and destinies
of the world’s cities are to be truly taken in hand. Yet the momentum is
growing and will surely prove unstoppable in the long run.
As was underlined by the UN Secretary-General’s High Level Panel on UN-
Civil Society Relations, the United Nations and its agencies, funds and pro-
grammes, such as UN-HABITAT, are at their most effective in promoting
positive change around the world when they reach out to the diverse range
of real actors on the ground and use their convening power to bring these
actors together to negotiate and agree on the way forward. In many
Vancouver, 1976, UNITED NATIONS
respects, the second United Nations Conference on Human Settlements,
held in Istanbul in 1996, broke new ground in the vision and practice of
partnership. In addressing the mounting challenges related to the huge
population shifts in the world’s cities, UN-HABITAT will need to build upon
and develop this legacy and deepen its relations and cooperation with a
wide array of governmental and civil society organizations everywhere that
are vital to its mission.
Tony Hill is the Coordinator of the United Nations Non-Governmental
Liaison Service (NGLS).
also the region where there has been the greatest upsurge in
slum growth rates since the start of the 1990s. In the coun-
tries such as Burkina Faso, Ethiopia, Senegal and Tanzania,
where decentralization initiatives are underway, the number
of slum dwellers has risen from between 4 per cent and 6.2
per cent per year on average. However, other countries are
performing better, including Ghana, with a slum growth rate
of 1.8 per cent per year. South Africa, in particular, stands
out in its efforts to keep slum growth rates down to only 0.2
per cent annually.
In this group of “reforming” countries, there does appear
to be some association; as a general rule, the more established
the local governance practices in a country, the more able a
country appears to be in managing its slum growth rates.
However, the relationship between good local governance
and its effect on reducing slum growth is often far from
clear-cut. For one thing, the move towards more decentral-
ization and local democratization does not seem to automat-
ically result in improvements in the lives of the urban poor,
especially over the short-term. With the possible exception of
South Africa, decentralization can, and often has, placed
power in the hands of local elites, particularly those that
played an established role under earlier, centralized systems,
and has simply worsened inequalities. Secondly, in most
parts of the developing world, especially sub-Saharan Africa,
decentralization is a relatively recent process and if experi-
ence of other regions is anything to go by, it will take decades
for decentralization to make an impact. And thirdly, gover-
nance alone cannot explain why slum growth rates have gone
up in some countries and down in others; economic develop-
ment, levels of urbanization, good and bad policies, all con-
tribute to a country’s overall performance.
It is nonetheless pertinent to ask: why does good local gov-
ernance make a difference in slum growth rates in some
places? Better local governance is starting to contribute to the
success of slum upgrading operations and their scaling up to
citywide and nationwide scales through various means. First,
decentralization brings with it new incentives for municipal
governments to participate in upgrading existing slums and
related poverty-reduction schemes – including the design,
implementation and financing of basic infrastructure and
services in poor neighbourhoods and informal settlements.
Second, municipal reforms and capacity building, in some
places, have led to the improved operation of local authori-
ties through changes in systems and administrative proce-
dures, streamlining of functions, and reform of municipal
financial systems. Enhancing the rule of law, efficiency in
service delivery, and fiscal transparency and accountability,
have been the major objectives of these processes. For exam-
ple, a key success factor behind the Favela Bairro pro-
grammes in Rio de Janeiro is that it is financed and executed
entirely by the municipality that has built up its own capac-
ity and institutional structures for efficient service delivery. A
technical committee drives investment decisions while a
coordination committee makes sure that all municipal
departments are on board for implementation. Furthermore,
the programme uses modern management techniques that
ensures smooth running of between 40 to 50 neighbourhood
projects at the same time all over the city. The programme is
also strong on ensuring community involvement in decision-
making and in the operation of some services such as garbage
collection and reforestation5. Third, new waves of democra-
tization and decentralization have also led to a marked
improvement in the organizational capacities of civil society
groups and opened up opportunities for slum dwellers and
other marginalized group, to get involved in planning and
project design and implementation. With growing political
maturity of grassroots organizations, the urban poor are in a
better position to negotiate with local authorities for servic-
es and land rights.
Local governance works, but in many countries it works best
with strong support from the centre. Countries that have per-
formed well in decentralizing and strengthening local gover-
nance have done so with strong commitment and support
from central government. National reforms and legislation –
for example, covering decentralization, fiscal transfers,
municipal elections, community participation, and spatial
planning – create the enabling environment for city admin-
istrations to carry out their functions, including slum
upgrading and prevention, more effectively. Despite hand-
ing over power, governments continue to play a significant
role in taking decisions from the top that invariably affect
the provision of shelter and services for the poor at the local
level. For example, while many Latin American countries
have demonstrated a good track record in decentralization,
reforms still often tend to emphasize de-concentration
rather than real redistribution of power. Decision-making
power, in many ways, remains centralised. In addition,
despite new systems of political representation and partici-
pation of civil society, countries in the region often display
centralized structures that serve to strengthen the position
of the ruling political party, reflecting the continuing cen-
tralist character of Latin American political culture.6 For
instance, in Mexico, the central government, through the
Ministry of Social Development (SEDESOL), is imple-
menting a large-scale national urban poverty reduction pro-
gramme – “Habitat Programme” – involving significant
transfers of resources to the local level. However, the
resources and activities are, in effect, managed by local
branches of the central administration or by local structures
reinforced by the central government through so-called
“local development agencies”. Despite working in close col-
laboration with the municipality and other local stakehold-
ers, control lies principally in the hands of the central gov-
ernment. The ministry can indeed play a very positive role
in coordinating resources and delivering services for the
poor, implementing redistributive policies that aim at bridg-
ing the gap between regions and cities. What is important is
to ensure that bottom-up approaches to governance connect
with top-down systems of decision-making.
173
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Highly centralized systems of governance
cannot benefit the urban poor if they are
being run by regressive, anti-poor political
leaders and inequitable policies.
Where top-down centralized governance also works
Moving to the other end of the spectrum of countries, an
interesting finding has emerged: among those countries that
have highly centralized structures of governance some also
appear to be performing well on the slum target. By taking a
tough policy stand on improving slum and housing conditions,
governments in these countries have been able to set up the
institutional arrangements, allocate important budgets, and
execute projects to effectively meet their targets and commit-
ments. And they have done so with limited involvement of
local authorities or communities in decision-making processes.
This is particularly the case in the countries of Northern Africa,
such as Egypt, Morocco and Tunisia, where local governments
have no real political or administrative power, and where most
decision-making is centralized. These are also the countries that
are experiencing low or negative slum growth rates and where
central governments continue to exert enormous control over
allocation of resources and decision-making. Cuba also per-
formed well, recording a slum growth rate of 0.7 per cent per
annum since 1990, making it one of the best performing coun-
tries in the Latin American and the Caribbean region.
What is clear, however, is that highly centralized systems of
governance cannot benefit the urban poor if they are being run
by regressive, anti-poor political leaders and inequitable poli-
cies. In all centralized countries that have performed well on
reducing slum growth rates, benevolent, progressive leadership
appears to be key to the success of slum upgrading pro-
grammes. For instance, Morocco’s centralist tradition has ben-
efited from the new monarch’s pro-poor stance towards
upgrading. In Cuba, the island’s greatest achievements have
been based on the government’s consistent commitment to the
principles of universality and equitable access that have defined
social and economic policy since the 1959 revolution. On the
other hand, the Government of Zimbabwe, through its nation-
wide eviction programme in 2005, demonstrates how top-
down rule can adversely affect the urban poor.
Why has centralized governance, contrary to current think-
ing, also produced meaningful results in these places? It appears
that command and control from the centre has often given
cohesiveness to the design and implementation of slum
upgrading projects. For instance, in Tunisia, a large number of
institutions are involved in slum upgrading, including the
Urban Upgrading and Renovation Agency (ARRU), the
Housing Bank, the Solidarity Bank, the municipalities and
some non-governmental organizations and community groups.
The entire operation is managed by ARRU, which acts on
behalf of the municipality. Through this approach, the pro-
174
gramme has succeeded in upgrading more than 250,000 hous-
ing units (18 per cent of the urban housing stock) benefiting
about 1.5 million people. In Morocco, slum upgrading is also
driven from the centre by the Ministry of Housing among oth-
ers, and has produced good results – between 1993 and 1999,
82 slum upgrading projects were implemented reaching nearly
99,000 households7. As these examples show, top-down planning
and implementation by strong central institutions can play a key
role in the success of slum upgrading projects by providing clear
purpose and direction, effective coordination, and institutional
capacity to achieve results.
While a direct, top-down approach to slum upgrading and
shelter provision has brought benefits in these countries, there is
a significant downside. Without meaningful decentralization
and participation, it becomes harder to motivate municipal gov-
ernments, civil society and citizens to take more control over the
processes that affect their material well-being and contribute to
development. In addition, participatory processes make possible
collective learning, promote social and institutional innovation
and facilitate different forms of inclusion. Centralized interven-
tions can be extremely effective in redressing inequalities, but
they come with a social cost. In Cuba, for example, the idea that
the government is in a better position to determine how hous-
ing construction resources are to be used has stifled community
and individual initiatives to improve housing, particularly when
these initiatives fall outside national priorities.
However, in some cases, governments are starting to realize
that top-down decision-making can only go so far. For instance,
the Government of Morocco is now trying to introduce greater
participation in the planning and implementation of slum
upgrading projects through a new concept called “Social Project
Control” and laws on promoting participation. In Cuba, self-
help community-driven construction – through government-
sponsored “microbrigades” – has been broadly implemented in
the last two decades, although the impact of these initiatives is
not yet fully felt in larger cities. In Egypt, too, there are signs of
change. The entire regulatory and policy framework, including
the constitution, is being examined and revised in a massive
effort to broaden democratic processes within the country.
Endnotes
1 Taylor 2000.
2 Souza 2002.
3 Pieterse 2000.
4 Halfani 2004.
5 Inter-American Development Bank 2004.
6 Pieterse 2000.
7 This information was gathered during UN-HABITAT’s research on slum
policies in 23 countries. Sources included telephone interviews and var-
ious country reports.
ANTOINE HEUTY AND SANJAY G. REDDY
SLUM IMPROVEMENT: TECHNICAL FIXES ARE NOT ALWAYS THE SOLUTION
The ultimate costs of achieving the Millennium Development Goals in cities
may well be either significantly underestimated or significantly overestimated
because most cost estimates are based on unjustified assumptions and weak
data. For example, existing models assume that the unit costs of required
interventions to achieve a Goal are fixed, even as the Goal is progressively
attained. However, there are strong reasons to believe that decreasing or
increasing marginal costs (economies and diseconomies of scale) may play an
important role in determining costs. For instance, in poor urban areas, those
to whom coverage of relevant services (such as piped water, electricity or tele-
phones) must be extended may be those who are most difficult to reach, for
geographical or social reasons. In this case, the initial cost of delivering serv-
ices may be high, as it may depend on the installation of infrastructure need-
ed to make service delivery possible. On the other hand, it can become pro-
gressively easier and cheaper to provide more of the same services when the
delivery network and infrastructure already exists or has previously been
installed. Which of these is the case may depend on local circumstances.
The UN Millennium Project’s estimates of the cost of achieving Millennium
Development Goal 7, target 11 (developed in a background paper prepared
for UN-HABITAT and the Millennium Project’s “Task Force on Improving the
Lives of Slum Dwellers”) are based on an approach that assumes away such
increases or decreases in marginal costs and also fails to address differences
in unit costs between and within countries (although it does distinguish
between investments in existing and in new settlements). It also fails to take
account of interdependencies in the attainment of different Goals: there may
be economies or diseconomies of scope that operate between distinct kinds
of services; the cost of expanding services of one kind may depend on the
extent to which services of another kind have already been expanded. For
example, it may be less costly to bring about improvements in child health if
children attend school. On the other hand, increases in child survival will
increase the number of children for whom schools must be provided, and
thereby increase the cost of achieving school enrolment objectives.
Interactions of this kind among distinct social development objectives are
numerous, but are largely neglected in current work on the costs of achieving
the Millennium Development Goals. Finally, the data required to assess the
baseline scenario of the Goals and to monitor their progress over time are at
present severely deficient. As a result, it is often not possible meaningfully
to judge either the extent of progress required or the costs of achieving the
required level of progress. The baseline data may also understate the scale
and depth of poverty, especially in urban areas. The Millennium Project’s
estimate of the total cost of meeting target 11 for developing countries ($113
billion for slum upgrading for 100 million existing slum dwellers, and $1176
billion for new sites construction for 700 million potential future slum
dwellers) may end up being quite inaccurate because it is premised on a
model in which these considerations hardly figure.
Our work (see “The Cost of Achieving the Millennium Development Goals:
What’s Wrong with Existing Analytical Models” on www.millenniumdevel-
opmentgoals.org) has demonstrated that the costs of ultimately achieving
the Goals are potentially significantly affected by the assumptions that are
made regarding unit costs and the nature and extent of economies of scope
and scale, within plausible ranges of variation.
Although we already know about some interventions that are likely to be
effective in enhancing human well-being, the solutions to a great many
other problems in urban areas are unknown. Technical fixes may not exist
for the most important problems in urban areas, which may require institu-
tional innovations, political will and more innovative governance structures
(such as participatory budgeting). The use of technocratic models is ulti-
mately unlikely to provide an adequate basis for improving the living condi-
tions of the poor in cities.
Even the most carefully constructed future scenarios are ultimately unlike-
ly to prove accurate, especially when projected far into the future.
Unpredicted shocks, whether at national, regional or global levels, are
sure to eventually undermine the accuracy of such forecasts. Examples of
significant shocks of this nature that have arisen in the past or may occur
in the future include new diseases (such as HIV/AIDS), disruptive large-
scale climatic events (such as the December 2004 Indian Ocean tsunami or
the recent hurricanes in Louisiana in the United States), and civil and
regional conflicts. Of course, those who frame analytical models do not
claim that the strategies that they recommend can be applied without
regard to changes in the circumstances of their application. However,
these models do not take note of the likelihood that such changes in cir-
cumstance will arise.
The unreliability of the informational base and the undue restrictiveness of
current approaches for planning to achieve the Millennium Development
Goals means that these approaches cannot serve adequately to guide poli-
cymakers in the poorly charted, uncertain and changing environment that
they actually face. The weaknesses of technocratic predictive models can
be mitigated but not overcome. The potential damage from the use of incor-
rect predictive models in decision-making is likely to be greater when such
models are applied to guide long-term decision-making – whether over the
level of resources to be raised or over how those resources will be
deployed. Inaccurate predictive models can eventually cause significant
misallocation of resources and errors in policy choice, which may make it
difficult or impossible to achieve the Goals.
An alternative approach to strategic planning should establish an institu-
tional framework for continuous informed policy choice by representative
decision-makers. The alternative approach to achieving the Millennium
Development Goals can be implemented through a process of periodic peer
and partner review, through which countries, regions and cities periodical-
ly formulate and review one another’s plans, with the broad involvement of
citizens and the support of experts. Participatory planning and budgeting
that promote context-sensitive judgment and learning from experience can
be important elements of such a strategy.
Antoine Heuty is Public Finance Economist (Poverty Group) at the United
Nations Development Programme (UNDP).
Sanjay G. Reddy is Professor of Economics at Barnard College and in the
School of International and Public Affairs, Columbia University.
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
4.5 Time for Bold Action: Scaling up Improvements Today,
Preventing Slums Tomorrow
■ Countries changing course in the new
millennium
W
ith some notable exceptions, the 1990s
presented a bleak picture of policy out-
comes in reducing shelter deprivations
and improving the lives of the urban poor.
In the absence of more effective, ambi-
tious policies and programmes, city conditions will deteriorate
rapidly in most of the “off track” and “at risk” countries described
in Chapter 1.3. The challenge is daunting, yet a few developing
countries have made remarkable leaps in improving the lives of
slum dwellers over the last decade and a half. These are not seis-
mic shifts by any stretch of the imagination; rather change has
been driven by patient, consistent policies and leadership over
time. Poor-performing countries, if they choose to, can take
inspiration and hope from fast-track countries. Their govern-
ments can make a decision to change course today by making a
serious commitment to slum improvement and implementing a
bold action plan to meet target 11. Alternatively, they can decide
to stay on the same path towards 2020 and watch the numbers
of people living in slums grow.
It is not too late – countries off track can get back on track with
the determination and foresight to introduce long-term planning
and reforms for achieving the ‘cities without slums’ target. The first
step for these countries is to take target 11 and the other Millennium
Development Goals and targets seriously and to mobilize political
will behind the Goals. As the report of the Millennium Project to the
UN Secretary-General, Investing in Development: A Practical Plan to
Achieve the Millennium Development Goals emphasizes:
“To enable all countries to achieve the Millennium Development
Goals, the world must treat them not as abstract ambitions but as prac-
tical policy objectives. The Goals are essential for transparency and
accountability, so it is important that they be taken literally since the pres-
sures in development policy push overwhelmingly for lower rather than
higher expectation. National governments and international donors, not
wanting to be held accountable for their role in poverty reduction, will
always want to water down the Goals – particularly if achieving them
requires increased budgetary commitments or major policy changes. In
many countries the Goals are deemed ‘unrealistic’ because they would
require dramatic progress.”
There are signs to indicate that, at least in some countries, govern-
ments are starting to take target 11 more seriously. Countries that
struggled during the last decade, such as Mauritania and Senegal, for
instance, are now showing political determination to make slum
upgrading a core business and are moving towards longer-term,
scaled-up slum policies. Other well performing countries from the
176
nineties are stepping up their actions. For instance, Brazil, Mexico,
South Africa and Thailand are carrying forward comprehensive,
national upgrading policies to deal with regularizing and improving
conditions in existing slums, as well as trying to plan ahead to avoid
future slum growth. Yet, most countries that performed poorly dur-
ing the 1990s continue to lag behind in making the political com-
mitment and reforms needed – our projections indicate a worsening
situation in all of these places. Can the governments of poor-per-
forming countries find a compelling enough reason to act now and
with the ambition needed to achieve target 11?
If the answer is yes, they would be well served taking on board the
major lesson learned by successful countries: success is driven by follow-
ing a two-pronged strategy for, one, scaling up improvements in existing
slums and, two, planning well ahead to provide better, alternative solu-
tions to avoid the spread of future slums. Countries such as Brazil,
Mexico, Colombia, South Africa, Thailand and Tunisia, have all
managed to successfully scale-up programmes for slum and informal
settlements upgrading and urban poverty reduction to countrywide
levels. In doing so, these programmes have resulted in a measurable
impact on national indicators of slum growth. One-off, local projects,
however successful, are usually incapable of making such a mark
without widespread replication and scaling up. Pilot projects provide
valuable test cases and, when they work, demonstrate the technical
and financial feasibility of providing better housing and services to the
urban poor. Many of today’s successful national slum upgrading proj-
ects, such as Indonesia’s Kampong Improvement Programme, began
life on a modest scale, covering a few neighbourhoods or a single city
and, with a proven track record, were expanded to national level.
Equally importantly, these countries have realized that the magnitude
of deficiencies in basic infrastructure, service and shelter provision for
slum communities today will fade into insignificance compared to
conditions in the next five, ten or fifteen years. Governments have,
therefore, taken a much longer view on the expansion of slums and
have begun to focus their energies on measures that effectively meet
future need by developing plans that can effectively halt the growth
of new slums and promote more sustainable cities. This has meant
creating a planning system that makes land and infrastructure avail-
able and affordable to low-income housing.
■ Towards national and local strategies to achieve
Millennium Development Goal 7, target 11
China UNEP/STILL PICTURES
Those countries that are serious about achieving target 11
should make sure they have a long-term, national strategic plan
for scaling up remedial and preventive measures to meet the
basic needs of slum dwellers and future low-income popula-
177
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
tions. The national plan would provide a blueprint for action
that signals the government’s commitment to improving the
lives of slum dwellers and sets out clear, time-bound targets and
new policy vehicles for achieving widespread governance and
sectoral reforms. The plan should include appropriate budgets
and expenditure frameworks for achieving its stated objectives
and targets. Such a planning process would offer countries an
opportunity to formulate a countrywide policy for urban
poverty reduction that is aligned systematically with all of the
Millennium Development Goals and targets. Brazil, for exam-
ple, has taken steps to shape new national policies responding
to target 11 and the achievement of other Goals and targets in
its cities. The Ministry of Cities, established in 2003, is cur-
rently responsible for formulating and implementing the new
National Urban Development Policy. The overarching policy
framework is based on the principle of universal access to ade-
quate housing, urban land, safe drinking water, sanitation and
mobility with safety.1
Likewise, municipal authorities should also be in the fore-
front of translating the Millennium Development Goals, par-
ticularly Goal 7, target 11, into their own city-level goals and
targets, and subsequently adopt citywide strategies for achiev-
ing them. The goals may be global in character but they must
be implemented locally, at city and community levels, where
people live and shelter and services are required. However,
many local governments are barely aware of target 11 and the
other Millennium Development Goal targets and even if they
are, they often have no incentive or commitment to meeting
these targets. Thus the first task before local governments is to
build awareness about the Goals, why they are important, and
what they mean to a slum dweller in the city.
Local governments are no strangers to setting targets – for
housing, infrastructure, services, health, education and other
sectors. The key to achieving the Millennium Development
Goals is: (a) to benchmark the targets against the Goals; and (b)
to ensure that the targets are bold enough to deal with current
shortfalls and are established in consultation with national gov-
ernments and local stakeholders. This would require many
rounds of consultation, discussion and explanation, sharing of
experiences and best practices. Such participatory processes can
build awareness among local authorities and stakeholders, and
demonstrate how the Goals are linked to their own objectives
and priorities. Local authorities should also try to produce bet-
ter information to give as accurate a picture as possible of the
situation – be it the number of people living in slums, the num-
bers without clean water and sanitation or the numbers of chil-
dren dropping out of school. This kind of detailed information
helps planners and policy-makers to make more informed deci-
sions and keep track of change.
Strategies and action plans to achieve the Goals must cut
across sectors and institutions. A common folly is to establish a
stand-alone local authority project or department for poverty
reduction that is separate and distinct from the project or
department that monitors the achievement of the Goals at the
national level. This separation defeats the very purpose of the
178
Goals, particularly Goal 7, target 11. The Millennium
Development Goals and targets must be built into all develop-
ment activities and projects, and resources must clearly be allo-
cated for these. Often, intervention in just one sector, such as
improving sanitation or regularizing tenure, can have a huge
impact on the quality of living conditions in slums. Targeted
interventions, aimed at the most vulnerable urban populations,
can sometimes be more effective than physically upgrading
slums, which may not be feasible or viable in the short-term. In
the regions that suffer from one major shelter deprivation, for
instance, intervention in just that sector could drastically
reduce the number of slum dwellers. For instance, in both
Tanzania and Uganda, where over 80 per cent of the urban
population suffers from lack of proper sanitation, investment
in sanitation in slum areas could reduce the proportion of slum
dwellers from more than 90 per cent to 40 per cent of the
urban population in both countries, assuming that the other
shelter needs, such as water or sufficient living area remain the
same. Egypt is one country that managed to dramatically
decrease slum incidence by investing heavily in water and san-
itation.
■ Mobilizing financing for pro-poor urban strategies
and programmes
Most of the investments needed to achieve target 11 and the
other Millennium Development Goals and targets in cities will
have to come from domestic sources. With limited budget sup-
port from the centre, cities need to turn to public sector borrow-
ing in the domestic financial markets to fund major investments
in infrastructure and services. However, most developing cities
have sourced debt financing mainly from government financial
institutions or on the basis of government guarantees. In this
regard, national policies and regulatory reforms can play a key
part in removing distortions in the market and attracting private
capital to finance public infrastructure2. Another crucial measure
to narrow the gap between municipal financial resources and
expenditure is to enhance the revenue base of the local authority,
for example, through increasing the efficiency of property tax
collection or rationalizing water rates and ensuring that revenue
collected is devoted to slum upgrading and prevention.
In addition, slum communities can make a major contribu-
tion to upgrading through their own savings and by leveraging
various sources of local funding. Innovative mechanisms
should be looked at to consider how best to improve access to
credit among the urban poor. For instance, housing microfi-
nance has been quite successful in reaching low-income groups.
Municipal subsidies are also being explored in different coun-
tries. However, what is perhaps most needed is long-term
finance for low-income shelter. This is a gap that the private
sector, non-governmental organizations and donors are cur-
rently trying to fill by supporting demonstration pilot projects
in which local authorities and civil society recipients can bor-
row funds for shelter development and slum upgrading.
■ Leveraging Local Resources for Slum Upgrading
International assistance towards housing and basic services in
developing countries has not been sufficient to address the sig-
nificant shortfall in these areas. It is estimated that combined
public and private investment and official development assis-
tance meets only 5 per cent to 10 per cent of the financing
required for slum upgrading in sub-Saharan Africa, Southern
Asia and South-Eastern Asia. In order to deal directly with this
“finance gap” in slum upgrading, investments must be predom-
inantly domestic, community-driven and market-based.
Any international donor interventions must therefore be cat-
alytic and should “leverage in” local and other resources. Many
developing countries have a large amount of resources that
could potentially be directed towards slum improvement – the
challenge is that such resources need to be harnessed, priori-
tized and restructured. The real barrier is the lack of political
will, accountability and institutional capacity, rather than
financial affordability.
The international donor community has an important role
in helping to develop sustainable financing mechanisms for
slum upgrading and prevention that should be built on the fol-
lowing three pillars:
(a) Harnessing and enhancing individual and community
resources;
(b) Strengthening and reallocating finances of city and nation-
al governments to meet the needs of the urban poor, and
the introduction of appropriate financial and non-financial
public policy instruments; and,
(c) Promotion of access to domestic capital markets.
There are increasing levels of community mobilization and
savings in slums. Simply put, slum dwellers are taking matters
into their own hands in the absence of affordable housing and
related urban infrastructure – and the absence of public and
private resources to finance such improvements. The result is
a proliferation of daily savings associations, work-based sav-
ings and credit schemes, revolving loan funds, and micro-
finance lending. While social lending arrangements of this
kind vary from slum to slum, city to city and country to
country, they share in common powerful mechanisms for
both mobilizing savings and undertaking community-based
initiatives to improve housing and infrastructure. The experi-
ences of community-led and government-enabled pro-
grammes in countries such as India, Morocco, Tunisia and
Uganda demonstrate the principles of a workable communi-
ty-driven approach.
There are also cases of countries, such as India, Indonesia,
South Africa and Sri Lanka that have restructured public
finances and market-based financing for urban upgrading.
However, the experiences so far have largely remained as “iso-
lated islands of innovation” in many countries and there is an
urgent need to replicate and scale up actions. The process of
scaling up will require massive policy and regulatory reforms
that promote community driven approaches, healthy local gov-
ernment financial capacity and domestic capital markets.
The liberalization of the domestic financial service industry
is a key trend that the international community should seek to
capture and harness for the purposes of financing slum upgrad-
ing. While Brazil, India, Mexico and South Africa have gone
through significant liberalization of the banking sector and
opening of domestic capital markets, a similar, largely un-
noticed trend is also unfolding in lower-income countries in
sub-Saharan Africa, Southern Asia and South-Eastern Asia.
Pension funds, insurance companies and private investors in
these countries maintain enormous stocks of domestic capital
(estimated annual value of domestic capital in Nairobi, Kenya,
for instance, is $1 billion) and are increasingly trading on local
stock exchanges.
Taking upgrading projects to scale also requires access to
multiple forms of investment and the use of several kinds of
corresponding financial instruments and products. In some
cases, “credit enhancement” may be needed to attract domestic
capital. There are a variety of institutions currently providing
different forms of credit enhancements for projects that seek to
access capital markets, including, for example, the
International Finance Corporation (IFC), GuarantCo, United
States Agency for International Development (USAID)
Development Credit Authority (DCA) Facility, and the
Emerging Africa Infrastructure Fund. Domestic guarantee
facilities are also gradually emerging in a few countries in order
to attract private capital, including Colombia, India and South
Africa.
One of the constraints in attracting private capital is lack of
adequate “bankable” projects that addresses the risks and con-
cerns of communities, governments and the private sector. This
is the reason why UN-HABITAT, in association with several
donor agencies and development partners, set up the Slum
Upgrading Facility (SUF). The central objective of SUF is to
assist developing countries to mobilize domestic capital for
slum and urban upgrading activities. A major focus of SUF will
be to package the different forms of investment and to struc-
ture the projects so that these can attract not one but multiple
forms of financing. This process of rendering projects “bank-
able” will involve facilitating partnerships and strengthening
capacity at country level among development partners and the
domestic financial service industry. It will also include linking
these local actors with key international financial institutions,
donor facilities, and regional development banks and funds
that will be in a position to “credit enhance” domestic financial
instruments through risk reduction and risk sharing and, by
doing so, enhance the mobilization of domestic capital into
slum upgrading projects.
Endnotes
1 Government of Brazil 2005.
2 Cities Alliance 2004.
179
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
4.6 Is the International Community Ready to Keep the Promise?
I
n order to improve the living conditions of people in
slums, responsibility for change ultimately lies with gov-
ernments. Some countries are starting to show real deter-
mination in taking on the target of improving the lives of
slum dwellers and are making it a reality within their
cities and towns. Yet too many governments remain in a state of
inaction. Time has come for governments to place the urban
agenda much higher on the list of national developments prior-
ities – by identifying local resources, mobilizing domestic capi-
tal and developing mechanisms to attract external funds for
innovative solutions that would maximize slum upgrading and
prevention programmes.
Lack of investment in slums bears enormous social and eco-
nomic costs, which add to the burden of cities and govern-
ments. Development assistance in improving the capacity of
governments – institutional reforms, better local governance,
improved urban planning and management and providing
affordable land and housing solutions to the urban poor – can
go a long way towards creating a pro-urban environment in
countries of the developing world that currently do not address
slum upgrading or prevention as part of their overall poverty
reduction strategies.
In addition, carrying forward commitments made by rich
countries in recent years, particularly during international con-
ferences, it is possible to arrive at a “new deal” as part of the
Millennium Development Goals. The international system –
the United Nations, bilateral donors, the World Bank, and the
regional development banks – has proven to be an important
source of financing for poverty reduction in several developing
countries around the world. For many countries in Africa and
in least developed countries, aid is still the largest source of
external financing and, it is argued, is critical to the achieve-
ment of the Millennium Development Goals and targets.1
With this in mind, making the international system work bet-
ter for poor countries, especially by raising the amount and
quality of aid, is another major target that the world has set for
itself and expressed in Goal 8: to develop a global partnership
for development.
■ A question of money, a question of donor
attitude
Donor financing has played an important role in supporting
slum upgrading over the last few decades. International financial
lending institutions, regional development banks and bilateral
donors have provided consistent support to slum projects in this
180
period. In many countries, especially in Africa, international
agencies have provided the bulk of financing for slum upgrad-
ing primarily through investments and loans. The volume of aid
for the urban sector has consistently risen among some interna-
tional actors. For example, the Inter-American Development
Bank, operating in Latin America – the most urbanized of
developing regions – has seen its portfolio of urban loans grow
in volume and complexity; more than $25 billion of loans has
gone towards urban projects in the last 40 years, representing
nearly 15 per cent of the total lending by the Bank.2
The “Cities without Slums” target3 has helped to generate a
renewed interest in slum improvement among many of the
donor agencies. The target has sent out a clear signal to donors
and governments alike to re-orient their policies towards urban
poverty reduction. Multilateral and bilateral development agen-
cies are taking up this challenge and starting to streamline their
assistance to respond directly to the slum target. To some extent,
this explicit support for the slum target has the potential of
translating into higher levels of development assistance targeted
specifically at slum upgrading and slum prevention.
Within this framework, UN-HABITAT has transformed its
work programme in line with target 11 as well as other
Millennium Development Goals and targets, including those on
water and sanitation. Slum upgrading is now an important area
of focus for the organization, with increasing emphasis being
placed on policy and operational support to the following areas:
scaling up of slum upgrading projects and programmes; cam-
paigns on secure tenure and urban governance; urban water sup-
ply and sanitation; and pro-poor planning and management.
Monitoring progress towards achievement of the slum target at
the national level is also an important part of UN-HABITAT’s
work. The agency is also leading a major new initiative – the
Slum Upgrading Facility (SUF) – designed to assist local part-
ners to mobilize local domestic capital for slum upgrading, low-
income housing and related infrastructure. The Cities Alliance,
a joint initiative of UN-HABITAT and the World Bank that
brings together a global coalition of cities and their development
partners, has also played a catalytic role in coordinating and
Kibera, Nairobi SEAN SPRAGUE/STILL PICTURES
mobilizing broad-based international support for scaling up
slum upgrading activities.
Despite this overall improvement, development assistance to
alleviate urban poverty and improve slums remains woefully
inadequate. Although investments have been made in related
sectors, such as health, water and sanitation, these are not tar-
geted specifically at slums, particularly in developing countries.
The cost of meeting the slum target alone – that is improving
the lives of at least 100 million slum dwellers – has been estimat-
181
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
ed at $67 billion4. Furthermore, providing decent housing and
basic services for the additional 400 million people who are
expected to join the ranks of slum dwellers during by 2020 is
estimated to cost a total $300 billion or over $20 billion per
year. Successful models have demonstrated that, when appropri-
ately supported by local and central governments, local residents
can provide about 80 per cent of the required resources. This
would leave 20 per cent to be provided by international aid, that
is roughly US$5 billion a year. Yet, according to one estimate,
total urban sector assistance to developing countries is just $2
billion a year, a fraction of what is needed to meet the slum tar-
get and cope with future growth5, that is, if all funds are used
for this purpose.
Although urban poverty is beginning to be recognized as an
issue that deserves attention, it is apparent that international
assistance to the urban sector has not been able to match the
scale of the problem. Associated with low levels of resource
mobilization, the real question is that urbanization has not been
fully understood by developing agencies both in terms of posi-
tive outcomes and negative externalities. The international pol-
182
bold reforms in land, housing, basic infrastructure, financing
and planning. International aid, in recent years, is more tar-
geted towards trying to meet these larger policy objectives. In
this way, multilateral and bilateral development partners have
gone for a more “programmatic” approach to urban develop-
ment projects, focusing, for example, on institutional devel-
opment, municipal management, provision of security of
tenure, and reform of central-local fiscal relations. This marks
a broad swing away from targeted area investments that
proved inefficient because of counterproductive policies,
especially concerning land regulation, and weak local institu-
tions that lacked the mandate and resources to deliver servic-
es on the ground8.
Yet, there is plenty of room for improving the strategic
focus of donor interventions in slum upgrading and slum pre-
vention. Firstly, despite increasing efforts to target policy fail-
ures at the national level and reinforce country leadership
behind slum improvement, more could be done in this area.
There is sometimes a tendency among governments and
donors to focus more on the local scene rather than dealing
head on with some of the basic weaknesses in the broader,
Rio de Janeiro from Corcovado Peak SEAN SPRAGUE/STILL PICTURES
national policy environment. Tackling the root causes – weak
institutions, stifling legal and regulatory systems – by putting
in place key reforms is often a much more difficult task. In
doing so, the international community is no longer providing
simply technical solutions but challenging governments to
make political choices. It takes courage for governments to
institute national reforms and programmes to benefit slum
dwellers because it means sharing power, wealth and land in a
more equitable way. The evidence suggests that countries,
such as Brazil, South Africa and Sri Lanka that have taken this
path have made major breakthroughs in reducing slum
growth rates.
Furthermore, while there is a deliberate attempt by donors
to “move purposefully to promote initiatives owned, generat-
icy environment needs to rethink the urban agenda and place it ed and designed by cities”9 in order to counter the dominant
among national priorities. This is the only way resources will power of central government that is often perceived to block
flow to address slum upgrading and prevent slum formation. city development, the evidence presented here suggests that
There are, however, some positive signs, particularly among the State, when it wants to, can play an extremely positive role
regional development banks; some governments are also revisit- in coordinating and delivering slum upgrading programmes.
ing, or at least debating about6, their aid programmes with the This perhaps calls for a further consideration of balancing
aim of integrating urban challenges in their development assis- support for both top-down and bottom-up governance
tance priority list.7 approaches. Poverty Reduction Strategy Papers (PRSPs) are a
good channel to do so, but the urban chapter needs to be pri-
oritized, confronting, somehow, the anti-urban attitude that is
■ Sharpening the focus of aid often found on both sides of the table (governments and
donors). Evidence shows that countries such as Jordan and
Another trend worth noting is that donors are becoming Tunisia that have successfully linked slum improvement ini-
more strategic in the way they support urban poverty reduc- tiatives to wider poverty reduction strategies have managed to
tion. As this chapter has highlighted, some countries have secure a bigger share of the national budget for slums and
achieved dramatic improvements in the living conditions of found it easier to push through reforms in key sectors. In most
the urban poor by hitting the most critical policy levers for of the other countries, governments and donors have not per-
scaling up and intensifying slum upgrading and prevention formed well in linking slum upgrading and urban poverty
measures, including building much-needed political leader- reduction to the broader, national development agenda. High
ship, enhancing local and central governance, and making slum growth rates in these countries prove this point.
It is time for governments and donors to take a more pro-
active stand in promoting Millennium Development Goal 7,
target 11 in their PRSPs – this may be one of the most critical
entry points for rapid scaling up of slum upgrading and slum
prevention measures. Yes, governments and donors, given their
limited resources, will need to balance competing priorities. But
the Goals and targets point them to a very clear set of priorities
and outcomes, improving health, education, the environment,
gender disparities and urban poverty issues. Hence, there is a
clear opportunity to generate momentum behind target 11 and
provide a commitment to long-term planning for slum growth
and prevention and connecting this with reforms in land, water
and decentralization policies.
But this takes the will of governments and donors to act – and
this is, for the most part, sorely lacking. The prioritization of
sectors and budget allocations by donors and recipient govern-
ment are often based on political decisions and are usually the
result of competing interests among different agencies and sec-
tors. This is unfortunate, since such a process can ignore real
need and fail to consider past performance. When the issue of
urban poverty and slums are already well below the radar screen
of donors and governments, it makes it all the harder to ensure
a place at the negotiating table. To gain more visibility, donors
may consider new tactics, for example, when it comes to the
development and execution of urban projects, donors tradition-
ally deal with the line ministries such as planning, housing,
environment and local government. Perhaps, they could also try
to influence other parts of the central government, including the
ministries of finance, health and education that have more clout
in the PRSP process. This Report has shown the very significant
links between shelter deprivations in slums and health, educa-
tion and employment – it makes little sense to put “slums” in a
box without taking into account the critical linkages with these
other key sectors.
Endnotes
1 United Nations 2002.
2 Inter-American Development Bank 2004.
3 Millennium Declaration, paragraph 19.
4 UN Millennium Project 2005a.
5 Cohen 2004.
6 For instance, the February 2006 discussion held at the United Kingdom
House of Commons about urban poverty issues.
7 Mauritania is one example.
8 Kessides 1997.
9 Cities Alliance 2004.
183
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Bibliography
A Millennium. Geneva: SDR Secretariat, Office for the Coordination of Humanitarian
Addis Ababa City Council (2004). Household survey results shared with UN-
HABITAT as part of the Monitoring Urban Inequities Programme.
African Population and Health Research Centre (2002, April). Population and
Health Dynamics in Nairobi’s Informal Settlements: Report of the Nairobi Cross-
Sectional Slums Survey 2000. Nairobi.
Améndola, G. & Garay, A. (2000). “La visión urbanística”. In La Fragmentación
Fisica de Nuestras Ciudades. Municipalidad de Malvinas Argentinas.
Appadurai, A. (1996). Modernity at Large: Cultural Dimensions of Globalization.
Minneapolis: University of Minnesota Press.
Applebome, P., Drew, C., Longman, J. & Revkin, A. (2005, 4 September). “A
delicate balance is undone in a flash, and a battered city waits”. The New York
Times.
Arputham, J. (2001). “Whose city is it anyway?” Our Planet. UNEP.
Asian Coalition for Housing Rights (2001, October). “Building an urban poor
people’s movement in Phnom Penh, Cambodia”. Environment and Urbanization,
13(2).
Atkinson, A. (1996). “Sustainable cities: dilemmas and options”. City, 3/4, 5-11.
Augustinas, C. (2003). “Land in an urbanizing world”. Habitat Debate, 9(4).
Australian Bureau of Statistics (2004, September). “The health and welfare of
Australia’s aboriginal peoples”. Shelter WA Newsletter.
Ayieko, F. (2006, 30 January-5 February). “Form a Building Authority, say experts”.
The East African.
B Centre on Housing Rights and Evictions (COHRE) (2004, December). Evictions
BBC News (2005a, 15 July). “Regions and territories: Somaliland”. Retrieved from
http://news.bbc.co.uk.
BBC News (2005b, 30 August). “Immigrants die in new Paris fire”. Retrieved 30
October 2005 from http://news.bbc.co.uk.
BBC News (2005c, 15 November). “Chirac in new pledge to end riots”. Retrieved
16 November 2005 from http://news.bbc.co.uk.
BBC News, World Edition (2005a, 14 January). “Green hydrogen buses are a ‘hit’”.
Retrieved 26 October 2005 from http://news.bbc.co.uk.
BBC News, World Edition (2005b, 23 February). “Kenya urged to ban plastic bags”.
Retrieved 16 November 2005 from http://www.bbc.co.uk.
Bartram, J., Lewis, K., Lenton, R., & Wright, A. (2005). “Focusing on improved
water and sanitation for health”. The Lancet, 365, 810-812.
Bazoglu, N. & Moreno, E. L. (2005). UN-HABITAT Methodology Paper to Globally
Monitor Secure Tenure. UN-HABITAT, unpublished document.
Bennhold, K. (2005, 30 August). “Seven die in Paris fire, the 2nd in 4 days”. The
New York Times.
Biau, D. (2005). “Spatial inequalities – the need for affirmative action”. Habitat
Debate, 11(2).
Bidinger, P.D., Nag, B. & Babu, P. (1986). “Nutritional and health consequences of
seasonal fluctuations in household food availability”. Food Nutrition Bulletin, 8,
36-59.
Billig, P., Bendahmane, D. & Swindale, A. (1999). Water and Sanitation Indicators
Measurement Guide. Washington, D.C.: Food and Nutrition Technical Assistance
Project.
Bindra, S. (2005, 24 April). “Give shade to jua kali businesses”. Sunday Nation.
[Nairobi, Kenya].
184
Boulle, P. & Palm, E. (2004). Strategies to Reduce Urban Risk in the New
Affairs. Retrieved November 2005 from www.nidm.net.
Burra, S. (2005, April). “Towards a pro-poor framework for slum upgrading in
Mumbai, India”. Environment and Urbanization, 17(1).
Bwayo, J., Plummer, F., Omari, M., Mutere, A., Moses, S., Ndinya-Achola, J.,
Velentgas, P., & Kreiss, J. (1994, 27 June). “Human immunodeficiency virus
infection in long-distance truck drivers in east Africa”. Archives of Internal
Medicine, 154(12).
C Journal of Epidemiology, 143(6).
Caldeira, P. R. (1996). Building Up Walls: The New Pattern of Spatial Segregation
in São Paulo. UNESCO Report ISSJ 147/1996.
Canada Mortgage and Housing Corporation (2005, 1 April). Canada’s UN-Habitat
Urban Indicators: Monitoring “The Habitat Agenda” and “The Millennium
Development Goals”. CMHC File 0981-178. Report prepared for UN-HABITAT.
Carael, M. & Allen, S. (1995). “Women’s vulnerability to HIV/STD in sub-Saharan
Africa: an increasing evidence”. In P. Makinwa & A. M. Jensen (Eds.), Women’s
Position and Demographic Change in Sub-Saharan Africa. Westport, CT:
Greenwood Press.
The Center for International Earth Science Information Network (2005, 8 August).
Earth Institute News. Columbia University. Retrieved September 2005 from
http://www.ciesin.org.
Central Bureau of Statistics (Kenya) (2000). Economic Survey. Government of Kenya.
Centre on Housing Rights and Evictions (COHRE) (2003, 23 June). “Seven million
left homeless as forced evictions double in two years”. [Media Release]. Retrieved
15 February 2006 from http://www.cohre.org.
Monitor, 1(2). Retrieved 15 February 2006 from http://www.cohre.org.
Centre on Housing Rights and Evictions (COHRE) (2005, 29 November). “China
named a 2005 Housing Rights Violator for widespread forced evictions”. [Media
Release]. Retrieved 15 February 2006 from http://www.cohre.org.
Chambers of Justice, Care Kenya & Cradle (2005). The Defilement Index.
Cherp, A. (1999). Water, Environment and Children in Central Asia. Paper prepared
for UNICEF. Retrieved December 2005 from
http://www.ceu.hu/envsci/aleg/projects/water.pdf.
Chibli, M. (2004). Aleppo Local Urban Observatory. Presentation made at the
second World Urban Forum, 13-17 September, Barcelona.
Cities Alliance (2003). 2003 Annual Report. Retrieved from
http://www.citiesalliance.org.
Cities Alliance (2004). 2004 Annual Report. Retrieved from
http://www.citiesalliance.org.
City of Vancouver (2001). CityPlan: Directions for Vancouver. Retrieved 17
November 2005 from http://vancouver.ca/commsvcs/planning/cityplan/dfvs.htm.
City of Vancouver (2004). City of Vancouver Planning Department Information
Sheet: CityFacts 2004. Retrieved 17 November 2005 from
http://vancouver.ca/commsvcs/planning.
Clichevsky, N. (2003). ECLAC, “Pobreza y acceso al suelo urbano. Algunas
interrogantes sobre las políticas de regularización en América Latina”. Medio
Ambiente y Desarrollo, Número 75. [Chile.]
Coaffee, J. (2004). “Recasting the ‘ring of steel’: designing out terrorism in the city
of London”. In S. Graham (Ed.), Cities, War and Terrorism. Malden, Oxford and
Carlton: Blackwell Publishers.
Cohen, M. (2002). Spiderman, the Twin Towers and Saskia Sassen. Unpublished
paper presented in Barcelona.
Cohen, M. A. (2004). Cities and the Wealth of Nations. Paper presented at a
parallel event held at the UNCTAD XI-URBIS meeting, June, São Paulo.
Cornia, A. G. & Menchini, G. (2005). The Pace and Distribution of Health
Improvements During the Last 40 Years: Some Preliminary Results. Paper
presented at the Forum on Human Development, 17 January, Paris.
Crombie, I. K, Irvine, L., Elliott, L. & Wallace, H. (2005). Closing the Health
Inequalities Gap: An International Perspective. Geneva: WHO Europe.
Dahl, R. (2004). “Vehicular manslaughter: the global epidemic of traffic deaths”.
Environmental Health Perspectives, 112(11). Retrieved 10 September 2005 from
http://ehp.niehs.nih.gov.
Daily Nation [Nairobi, Kenya]. (2005, 2 December). “AIDS infection rate declines”.
D adopt use of biodiesel fuel”. [Media Release]. Retrieved 26 October 2005 from
Darrundono, P. (2005). “The first large-scale upgrading programme: an Indonesia
success story”. Habitat Debate, 11(3).
Davis, K. (1951). The Population of India and Pakistan. Princeton: Princeton
University Press.
Davis, M. (2004, March-April). “Planet of slums”. New Left Review.
DeParle, J. (2005, 4 September). “What happens to a race deferred”. The New
York Times.
De Soto, H. (2000). The Mystery of Capital. London: Black Swan.
Department for International Development (DFID) (2005, March). Fighting
Poverty to Build a Safer World: A Strategy for Security and Development.
London: DFID.
Dodoo, F. N., Sloan, M. & Zulu, E. M. (2003). “Space, context and hardship:
socializing children into sexual activity in Kenyan slums”. In S. Agyei-Mensah & J.
B. Casterline (Eds). Reproduction and Social Context in Sub-Saharan Africa.
Westport, CT: Greenwood Press.
Dooley, E. (2002). “Canada restricts cars”. Environmental Health Perspectives,
110(7). Retrieved 10 September 2005 from http://ehp.niehs.nih.gov.
Dooley, E. (2004). “Death toll of biomass burning”. Environmental Health
Perspectives, 112(6). Retrieved 10 September 2005 from http://ehp.niehs.nih.gov.
Dooley, E. (2005). “Africa afire”. Environmental Health Perspectives, 113(9).
Retrieved 10 September 2005 from http://ehp.niehs.nih.gov.
Dwyer, G. (2005). Violence and the Poor. Asian Development Bank. Retrieved
December 2005 from http://www.adb.org.
E Edition. Retrieved 26 October 2005 from http://news.bbc.co.uk.
Eaton, L., Flisher, A. J., & Aaro, L. E. (2003). “Unsafe sexual behaviour in South
African youth”. Social Science Medicine, 56, 149-165.
Economic Commission for Africa (2002). Youth and Employment in Africa. Paper
presented at the Youth Employment Summit, 7-11 September, Alexandria, Egypt.
Economic Commission for Latin America and the Caribbean (ECLAC) (2005). El
Rostro de la Pobreza en las Ciudades de América Latina y el Caribe. Report co-
published with UN-HABITAT.
Economic and Social Commission for Asia and the Pacific (ESCAP) (2003). Asia-
Pacific Economies: Resilience in Challenging Times. Economic and Social Survey of
Asia and the Pacific, 2003. New York: UN ESCAP.
Economic and Social Commission for Western Asia (ESCWA) (2001). Sustainable
Urban Development: A Regional Perspective on Good Urban Governance. New
York: United Nations.
The Economist (2002, 5 September). “Driven to Alcohol.” Retrieved 26 October
2005 from http://www.economist.com.
Economist Intelligence Unit (2005, 3 October). “Vancouver tops liveability ranking
according to a new survey by the Economist Intelligence Unit.” [Media Release]
Retrieved 10 November 2005 from http://store.eiu.com.
Economy, E. (2005, 5 December). “The lessons of Harbin”. TIME.
Ellaway, A. & Macintyre, S. (1998). “Does housing tenure predict health in the UK
because it exposes people to different levels of housing related hazards in the
home or its surroundings?” Health & Place, 4(2), 141-150.
Eltayeb, G. E. (2003). “The case of Khartoum, Sudan”. In Understanding Slums:
Case Studies for the Global Report on Human Settlements 2003: The Challenge
of Slums. Development Planning Unit, University College London/UN-HABITAT.
Energy for Sustainable Development Limited (2003). Fuel Substitution: Poverty
Impacts on Biomass Fuel Suppliers. Final Technical Report to the UK Department
for International Development, KaR Project R8019.
Esray, S. (1996). “Water, waste and well-being: a multi-country study”. American
European Environment Agency. (2004). Air Pollution in Europe 1990-2000. Topic
report No. 4/2003. Copenhagen: EEA.
Evans, B. (2005). Securing Sanitation – The Compelling Case to Address the
Crisis. A report commissioned by the Government of Norway as input to the
Commission on Sustainable Development and its 2004-2005 focus on water,
sanitation and related issues. Stockholm. Retrieved December 2005 from
http://www.siwi.org.
F Context. Paper commissioned by the Royal Norwegian Ministry of Environment.
Fafo Institute for Applied International Studies (2003, May). “Living in Port-au-
Prince”. A presentation prepared for the Habitat Urban Indicators Workshop,
Nairobi.
Faroohar, R. (2005, 21 November). “It’s about jobs”. Newsweek.
Fernandes, E. & Varley, A. (Eds.) (1998). Illegal Cities: Law and Urban Change in
Developing Countries. London and New York: Zed Books Ltd.
Food and Agriculture Organization of the United Nations (2002). “FAO methodology
for estimating the prevalence of undernourishment”. In Proceedings of the
International Scientific Symposium on Measurement and Assessment of Food
Deprivation and Under-Nutrition. Rome: FAO.
Ford, P. (2005, 2 September). “Fires reveal housing crunch”. Christian Science
Monitor. Retrieved 5 September 2005 from http://www.csmonitor.com.
Forum on the Global South (2003). Anti Poverty or Anti Poor? The Millennium
Development Goals and the Eradication of Extreme Poverty and Hunger. A study
commissioned by UN Economic and Social Commission for Asia and the Pacific,
Bangkok.
Fouchard, L. (2003). “The case of Ibadan, Nigeria”. In Understanding Slums: Case
Studies for the Global Report on Human Settlements 2003: The Challenge of
Slums. Development Planning Unit, University College London/UN-HABITAT.
Franks, T. (2005, 24 October). “Cows make fuel for biogas train”. BBC News, UK
Fry, S., Cousins, B., & Olivola, K. (2002). Health of Children Living in Urban Slums
in Asia and the Near East: Review of Existing Literature and Data. Environmental
Health Project, Activity Report 109. USAID, Washington, D.C.
G
Girardet, H. (2004). Cities People Planet: Liveable Cities for a Sustainable World.
Chichester, UK: John Wiley and Sons.
Glasze, G. Webster, C.J. & Frantz, K. (Eds.). (2005). Private Cities: Local and Global
Perspectives. London: Routledge.
Gonzalez, D. (2005, 2 September). “From margins of society to center of the
tragedy”. The New York Times.
Government of Brazil (2005). Millennium Development Goals: Brazilian Monitoring
Report. Unpublished document.
Government of Chile (2004, December). Report to UN-HABITAT by the Minsterio de
Vivienda y Urbanismo on Chile’s progress toward Millennium Development Goal 7,
target 11. Unpublished document.
Government of Indonesia (2005, October). Settlement, Drinking Water and
Sanitation: Best Practices. Jakarta: Ministry of Works.
Government of Kenya/UNCHS (2001). Nairobi Situation Analysis Consultative
Report: Collaborative Nairobi Slum Upgrading Initiative. Nairobi.
Goytia, C. (2005, 10 May). “Titling views from Argentina”. Message sent to mailing
list of Geoffrey Payne and Associates, http://www.gpa.org.uk.
Greater New Orleans Community Data Center (2005). Orleans Parish: Income and
Poverty. Retrieved 6 September 2005 from http://www.gnocdc.org.
H
Halfani, M. (2004). Overview of Decentralization Policy Issues. Presentation made
on behalf of UN-HABITAT at the Inaugural Meeting of AGRED (Advisory group of
Experts on Decentralization), 9-10 March, Gatineau, Canada.
Halifax Regional Municipality (2004, 12 October). “Entire metro transit bus fleet to
http://www.halifax.ca.
Hallman, K. (2004). Socioeconomic Disadvantages and Unsafe Sex Behaviour
Among Young Women and Men in South Africa. Policy Research Paper No. 190.
New York: The Population Council.
Hansen, S. & Bathia, R. (2004, March). Water and Poverty in a Macro-Economic
Hardoy, J. E., Mitlin, D., & Satterthwaite, D. (2001). Environmental Problems in an
Urbanizing World. London: Earthscan.
Hargeisa Municipality (Somalia). (2003). Hargeisa Municipality Statistical
Abstract.
Hiasat, Y. (2005). “Jordan’s commitment to countrywide slum upgrading”. Habitat
Debate, 11(3).
Hinrichsen, D. (2002). “Singapore – the global city”. People & the Planet. Retrieved
12 November 2005 from http://www.peopleandplanet.net.
Housing Statistics Indonesia (Biro Pusat Statistik) (2000). “Statistics of permanent
and non-permanent houses in Indonesia”. Central Bureau of Statistics document.
Huchzermeyer, M. (2006). Slum Upgrading Initiatives in Kenya Within the Wider
Housing Market: A Housing Rights Concern. Discussion Paper No. 1/2006, Kenya
Housing Rights Project, COHRE Africa Programme Center on Housing Rights and
Evictions, Geneva.
I settlements”. Links. Retrieved 15 December, 2005, from www.oxfam.org.uk.
Inter-American Development Bank (2004, March). The Challenge of an Urban
Continent. Washington, D.C.
International Energy Agency. (1996). World Energy Outlook. Paris: IEA/OECD.
International Labour Organization (2002). “Women and men in the informal
economy: a statistical picture”. Retrieved November 2005 from
http://www.eldis.org.
International Labour Organization (2004). Global Employment Trends for Youth.
Geneva: International Labour Organization.
IRIN News (2005, 6 October). “Zimbabwe: access to treatment a concern for
displaced living with AIDS”. United Nations Office for the Coordination of
Humanitarian Affairs (OCHA). Retrieved 10 February 2006 from
http://www.irinnews.org.
J
Jonsson, Å. & Satterthwaite, D. (2000). Income-Based Poverty Lines: How well do
the levels set internationally and within each country reflect (a) the cost of living
in the larger/more prosperous/more expensive cities; and (b) the cost that the
urban poor pay? Paper prepared for the Panel on Urban Population Dynamics,
Committee on Population, National Research Council/National Academy of
Sciences. Washington, D.C.
K
Kantai, P. (2003). “Dandora burning”. EcoForum, 26(3), 24-31.
Karasov, C. (2000). “On a different scale: putting China’s environmental crisis in
perspective”. Environmental Health Perspectives, 108(10). Retrieved 10 September
2005 from http://ehp.niehs.nih.gov.
Kathuria, S. & Oberai, R. (2004, March). How to Curb Crime and Its Impact on
Private Sector Growth. Moderated online discussion, archived on
http://rru.worldbank.org/Discussions/Topics/Topic32.aspx.
Kessides, C. (1997). World Bank Experience with the Provision of Infrastructure
Services for the Urban Poor: Preliminary Identification and Review of Best
Practices. Washington, D.C.: World Bank.
Kirby, A. (2004, 13 December). “Pollution: a life and death issue”. BBC News.
Retrieved November 2005 from http://news.bbc.co.uk.
Kothari, M. (2005). Cited in Habitat Debate, 11(2).
L
Landman, K. (2003). Alley-Gating and Neighbourhood Gating: Are They Two Sides
of the Same Face? Paper delivered at the conference, Gated Communities: Building
Social Division or Safer Communities? 18-19 September, Glasgow.
Langley, E. (2002). The Changing Visage of French Housing Policy and Finance: A
Half-Century of Comprehensive, Complex and Compelling Home Building. Joint
Center for Housing Studies, Harvard University, Kennedy School of Government.
Retrieved 15 September 2005 from http://www.jchs.harvard.edu.
Leaf, M. (2005). “Vancouver, Canada: multicultural collaboration and
mainstreaming”. In M. Balbo (Ed.), International Migrants and the City. Nairobi:
UN-HABITAT.
Leavitt, M. (2000). Melting Pot. Greater New Orleans Community Data Center.
Retrieved 6 September 2005 from http://www.gnocdc.org.
Lee, L. & Ghanime, L. (2004, September). Millennium Development Goal 7
Summary Review, 67 Country MDG Reports. Country Reporting on MDG7, Ensuring
Environmental Sustainability (2nd ed.). UNDP.
Linchu, Z. & Zhi, Q. (2003). “The case of Shanghai, China”. In Understanding Slums:
Case Studies for the Global Report on Human Settlements 2003: The Challenge of
Slums. Development Planning Unit, University College London/UN-HABITAT.
London Research Centre, Housing and Research Section (2005, September).
Indicator 15: Housing Overcrowding - Living at Over 1 Person Per Room. Brent
Council. Retrieved from http://www.brent.gov.uk.
Lugano, E. & Sayer, G. (2003, October). “Girls’ education in Nairobi’s informal
Luthra, D. (2005, 25 June). “Sri Lanka’s slow tsunami response”. BBC News.
M
Manase, G., Mulenga, M. & Fawcett, B. (2001). Linking Urban Sanitation Agencies
with Poor Community Needs. Paper prepared for the 27th WEDC Conference,
Lusaka, Zambia. Retrieved December 2005 from http://wedc.lboro.ac.uk.
Marcuse, P. (2001). “Dangers posed by ‘really existing globalization’”. Habitat
Debate, 7(4).
Martínez-Martín, J. (2005). Monitoring Intra-Urban Inequalities with GIS-Based
Indicators: With a Case Study in Rosario, Argentina. ITC Dissertation Series No.
127, Thesis, Utrecht University and ITC.
Martinot, E. (2005a). Renewables 2005: Global Status Report. Washington, D.C.:
Worldwatch Institute.
Martinot, E. (2005b). “Solar (and sustainable) cities”. Retrieved 26 October 2005
from http://www.martinot.info/solarcities.htm.
McMahon, J. M., Tortu, S., Pouget, E. R., Hamid, R., & Torres, L. (2004). Increased
Sexual Risk Behaviour and High HIV Seroincidence Among Drug-Using Low-Income
Women with Primary Heterosexual Partners. Paper presented at the 15th
International AIDS Conference, 11-16 July, Bangkok.
185
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Mehta, S. (2004). Maximum City: Bombay Lost and Found. London: Review.
Mendes, W. & Holden, M. (2005). Vancouver: Promoting Multiculturalism and
Social Inclusion Through Policy and Planning. Paper commissioned by UN-HABITAT
for this Report.
Milanovik, B. (2005, 21 November). “World apart”. Newsweek.
Ministerio de Vivienda y Urbanismo (2004). Metas Del Milenio. Meta Nº 11,
Informe del Gobierno de Chile, Diciembre.
Mitullah, W. (2003). “The case of Nairobi, Kenya”. In Understanding Slums: Case
Studies for the Global Report on Human Settlements 2003: The Challenge of
Slums. Development Planning Unit, University College London/UN-HABITAT.
Moreau, R. & Mazumdar, S. (2005, 25 April). “Bombay dreams”. Newsweek.
Moreno, E. L. (2005). “How far is the world from the slum target?” Habitat Debate,
11(3).
Moser, C. (2005). “City violence and the poor”. In In Focus: Poverty and the City.
Brasilia: UNDP/International Poverty Centre.
Moser, C. & Rodgers, D. (2005, March). Change and Violence in Non-Conflict
Situations. Working Paper 245. London: Overseas Development Institute.
Mosley, W.H. & Chen, L. (1983). “Child survival: strategies for research”.
Population and Development Review, Supplement to Vol. 10.
Mougeot, L. J. (Ed.) (2005a). AGROPOLIS: The Social, Political and Environmental
Dimensions of Urban Agriculture. London and Ottawa: Earthscan-IDRC.
Mougeot, L. J. (2005b). Ecological Footprint and Urban Agriculture. Paper
commissioned by UN-HABITAT for this Report.
Mukherjee, N. (2001). Achieving Sustained Sanitation for the Poor. Water and
Sanitation Programme for East Asia and the Pacific. Jakarta, Indonesia.
Musa, P. (2005, 17 April). “Sewage vegetables flood city markets”. Daily Nation.
[Nairobi, Kenya]
Myers, D. & Baer, W. (1996). “The changing problem of overcrowded housing”.
Journal of the American Planning Association, 62, 66-84.
N Habitat Debate, 11(3).
Nalo, D. S. (2002). Report of the Joint CBS/UN-HABITAT Fieldwork for the
Delineation of the Urban Slums/Informal Settlements: A Case Study of Nairobi.
Unpublished technical report.
Narayan, D. (2000). Voices of the Poor - Can Anyone Hear Us? Washington, D.C.,
World Bank.
National Institute of Standards and Technology, Partnership for Advancing
Technology in Housing (1999). Improving Durability in Housing. Background paper
for the National Forum on Durability Research, Maryland, USA.
Netherlands Department for Housing (2005). Dutch Urban Indicators Questionnaire
of Amsterdam and Den Haag. Unpublished document produced for UN-HABITAT.
Nicolae, V. (2005). “Prejudice, suspicion and discrimination – the plight of Europe’s
Roma communities”. Habitat Debate, 11(2).
Nitti, R. & Sarkar, S. (2003, 7 January). “Reaching the poor through sustainable
partnerships: the slum sanitation program in Mumbai, India.” Urban Notes: Thematic
group on services to the urban poor. Washington D.C.: World Bank. Retrieved
October 2005 from http://www.worldbank.org/urban/upgrading/urban-notes.html.
Norris, M. & Shiels, P. (2004). Regular National Report on Housing Developments
in European Countries: Synthesis Report. Department of the Environment, Heritage
and Local Government, Ireland.
O York: Penguin.
O’Higgins, N. (2002). Government Policies that Promote Youth Employment. Paper
prepared for the World Youth Summit on Government Policy and Youth
Employment, 7-11 September, Alexandria, Egypt.
Orc Macro (2004). Demographic Health Surveys (1990-2003). Calverton, Maryland:
Orc Macro.
186
P Development for Water and Sanitation in Urban Areas. WSSCC, Geneva. Retrieved
Perlman, J.E. (2005). “The chronic poor in Rio de Janeiro: What has changed in
30 years?” In M. Keiner, M. Koll-Schretzenmayr & W.A. Schmid (Eds.), Managing
Urban Futures: Sustainability and Urban Growth in Developing Countries.
Aldershot and Burlington, VT: Ashgate.
Pettifor, A. E., Rees, H.V., Steffenson, A., Hlongwa-Madikizela, L., MacPhail, C.,
Vermaak, K., & Kleinshmidt, I. (2004). HIV and Sexual Behaviour Among Young
South Africans: A National Survey of 15-24 Year Olds. Johannesburg:
Reproductive Health Research Unit, University of the Witwatersrand.
Philips, T. (2005, 29 October). “Blood, sweat and fears in favelas of Rio”. The
Guardian.
Phombeah, G. (2005, 2 September). “Living amidst the rubbish of Kenya’s slum”.
BBC News web feature. Retrieved 4 September 2005 from
http://www.bbcnews.co.uk.
Pieterse, E. (2000). Participatory Urban Governance: Practical Approaches,
Regional Trends and UMP Experiences. Urban Management Programme
Discussion Paper 25. Nairobi: UNCHS (Habitat)/UNDP/World Bank.
Planning Commission of India (2002). National Human Development Report 2001.
New Delhi.
Presidency of the Republic, Government of the Federative Republic of Brazil
(2004). Brazilian Monitoring Report on the Millennium Development Goals.
Institute for Applied Economic Research (IPEA) and National Institute of
Geography and Statistics (IBGE). Unpublished document.
Q T
Quan, J., Tan, S. F., & Toulmin, C. (Eds.) (2005, October). Land in Africa: Market
Asset or Secure Livelihood? Proceedings and Summary of Conclusions from the
Land in Africa Conference held in London, November 8-9, 2004. London: IIED.
Retrieved 15 February 2006 from http://www.iied.org.
R Teather, D. (2005, 5 September). “Neighbouring states struggle to cope with influx
Rakodi, C., Gatabaki-Kamau, R., & Devas, N. (2000). “Poverty and political conflict
in Mombasa”. Environment & Urbanization, 12(1).
Rakodi, C. (2005). “The urban challenge in Africa”. In M. Keiner, M. Koll-
Schretzenmayr & W.A. Schmid (Eds.), Managing Urban Futures: Sustainability
and Urban Growth in Developing Countries. Aldershot and Burlington, VT:
Ashgate.
Ravallion, M. (2001, April). On the Urbanization of Poverty. World Bank
Development Research Group, Policy Research Working Paper 2586.
Ray, K. (2003). “Water and sanitation in cities: translating global goals into local
action”. Habitat Debate, 9(3).
Risbud, N. (2003). “The case of Mumbai, India”. In Understanding Slums:
Case Studies for the Global Report on Human Settlements 2003: The
Challenge of Slums. Development Planning Unit, University College
London/UN-HABITAT.
Robertshaw, R., Louw, A., Shaw, M., Mashiyane, M., & Bretell, S. (2001, August).
Reducing Crime in Durban: A Victim Survey and Safer City Strategy. ISS
Monograph Series, No. 58.
Ross, N. A., Nobrega, K., & Dunn, J. (2001). “Income segregation, income
inequality and mortality in North American metropolitan areas”. GeoJournal,
53(2), 117-124.
S Retrieved 20 December 2005 from
Sachs, J. (2005). The End of Poverty: Economic Possibilities for Our Time. New
Satterthwaite, D. (1999). The Earthscan Reader in Sustainable Cities. London:
Earthscan.
Satterthwaite, D. (2004). The Underestimation of Urban Poverty. Poverty Reduction
in Urban Areas Series, Working Paper 14. London: International Institute for
Environment and Development.
Satterthwaite, D., McGranahan, G., & Mitlin, D. (2005, April). Community-Driven
October 2005 from http://www.wsscc.org.
Satterthwaite, D. & Tacoli, C. (2003). The Urban Part of Rural Development: The
Role of Small and Intermediate Urban Centres in Rural and Regional Development
and Poverty Reduction. International Institute for Environment and Development,
Rural-Urban Briefing Papers, No. 8. Retrieved November 2005 from
http://www.iied.org.
Scaramella, M. (2003). “The Case of Naples, Italy”. In Understanding Slums: Case
Studies for the Global Report on Human Settlements 2003: The Challenge of
Slums. Development Planning Unit, University College London/UN-HABITAT.
Secretariat of the Third World Water Forum (2003). Water Voice. Retrieved
September 2005 from http://www.worldwatercouncil.org.
Sharma, K. (2005a, 24 January). “Mumbai’s tragedy”. The Hindu.
Sharma, K. (2005b, 22 February). “Forget Shanghai, remember Mumbai”. The
Hindu.
Shisana, O. (2002). Nelson Mandela/HSRC Study of HIV/AIDS: South African
national HIV Prevalence, Behavioural Risks and Mass Media. Household Survey
2002. Cape Town: Human Sciences Research Council.
Souza, C. (2002). “Is the participatory budgeting process in Brazil over-rated?”
Habitat Debate, 8(1).
Statistics Canada (2001). Statistics retrieved 17 November 2005 from
http://www.statcan.ca.
Taylor, P. (2000). “UNCHS (Habitat) – the global campaign for good urban
governance”. Environment & Urbanization, 12(1).
Taylor, P. J. (2005). “Leading world cities: empirical evaluations of urban nodes in
multiple networks”. Urban Studies, 42(9), 1593-1608.
of people”. Guardian.
Tebbal, F. (2005). “Are countries working effectively on the Millennium targets?”
Thompson, J., Porras, I. T., Wood, E., Tumwine, J. K., Mujwahuzi, M. R., Katui-
Katua, M., & Johnstone, N. (2000). “Waiting at the tap: changes in urban water
use in East Africa over three decades”. Environment & Urbanization, 12(2), 37-52.
TIME (2005, 27 June). “A new China rising”.
U
UN-HABITAT (1995). Crowding and Health in Low Income Settlements. Nairobi:
UN-HABITAT.
UN-HABITAT (2001). The State of the World’s Cities Report 2001. Nairobi: UN-
HABITAT.
UN-HABITAT (2002a). “Livable Region Strategic Plan (LRSP) for the Greater
Vancouver Regional District”. Best Practices Database. Retrieved 17 November
2005 from http://www.bestpractices.org.
UN-HABITAT (2002b). Report of the Expert Group Meeting on “Urban Indicators:
Slums, Secure Tenure and Global Sample of Cities”. Unpublished report.
UN-HABITAT (2002c, 20 March). The Role of Cities in National and International
Development. HSP/WUF/1/DLG.II/Paper 2. Nairobi.
UN-HABITAT (2003a). The Challenge of Slums: Global Report on Human
Settlements. London and Sterling, VA: Earthscan.
UN-HABITAT (2003b). Guide to Monitoring Target 11: Improving the Lives of 100
Million Slum Dwellers. Nairobi: UN-HABITAT.
UN-HABITAT (2003c). Slums of the World. Nairobi: UN-HABITAT.
UN-HABITAT (2003d). Water and Sanitation in the World’s Cities. London and
Sterling, VA: Earthscan.
UN-HABITAT (2004a). Monitoring Urban Inequities Programme. Unpublished
concept paper by UN-HABITAT.
UN-HABITAT (2004b). Report of the Urban Inequities Survey in Addis Ababa.
Unpublished document.
UN-HABITAT (2004c). Urban Land for All. Nairobi: UN-HABITAT.
UN-HABITAT (2004d, December), Progress Report On Removing Discrimination
Against Women in Respect of Property and Inheritance Rights. Tools on
Improving Women’s Secure Tenure Series 1, Nr. 2. Nairobi: UN-HABITAT.
UN-HABITAT (2005a). Good Policies and Enabling Legislation for Attaining the
Millennium Development Goals. Nairobi: UN-HABITAT.
UN-HABITAT (2005b). Operational Activities Report 2005. Nairobi: UN-HABITAT.
UN-HABITAT (2005c). Urban Indicators Programme, Data Base III. Data from
2003. Unpublished database, Monitoring Systems Branch.
UN-HABITAT/DFID (2002). Sustainable Urbanization: Achieving Agenda 21.
Nairobi: UN-HABITAT/DFID.
UN Millennium Project (2005a). A Home in the City: Task Force Report on
Improving the Lives of Slum Dwellers. London and Sterling, VA: Earthscan.
UN Millennium Project (2005b). Investing in Development: A Practical Plan to
Achieve the Millennium Development Goals. Retrieved from
http://www.unmillenniumproject.org.
UN Office of the High Commissioner for Human Rights & UN-HABITAT (2002,
October). Global Housing Rights Challenge: Monitoring and Evaluation of
Progress. Geneva.
UNAIDS (2004). Report on the Global AIDS Epidemic. Geneva.
UNAIDS/WHO (2004a). Epidemiological Fact Sheet – 2004 Update: Kenya.
UNAIDS/WHO (2004b, December). AIDS Epidemic Update. Geneva: UNAIDS.
UNDP (2000). Azerbaijan Human Development Report 2000. Retrieved 16
February 2006 from http://www.un-az.org.
UNEP (2003). GEO Year Book, Freshwater: Meeting Our Goals, Sustaining Our
Future. Retrieved from http://www.unep.org.
UNEP (2005). Rapid Environmental Assessment of the Tsunami. Retrieved from
http://www.unep.org.
UNESCO (2003). Facts and Figures - Water and Cities. 2003 International Year of
Fresh Water Web site: http://www.wateryear2003.org.
UNFPA (2003). Achieving The Millennium Development Goals. New York.
Retrieved September 2005 from http://www.unfpa.org.
UNICEF (1997). Progress of Nations: Sanitation League Table. New York: UNICEF.
UNICEF (2003). Progress Since the World Summit for Children. New York:
UNICEF.
UNICEF (2004). The State of the World’s Children 2005: Children Under Threat.
New York: UNICEF.
UNICEF (2005). Progress for Children: A Report Card on Gender Parity and
Primary Education. (No.2). New York: UNICEF.
United Kingdom (2005). “The impact of overcrowding on health and education”.
Retrieved September 2005 from http://www.odpm.gov.uk.
United Kingdom Parliamentary Office of Science and Technology. (2002,
December). Access to Sanitation in Developing Countries. POSTnote No. 190.
http://www.parliament.uk/parliamentary_offices/post/environment.cfm#2002.
United Nations (2002). Monterrey Consensus of International Conference on
Financing for Development.
United Nations (2005a). The Inequality Predicament: Report on the World Social
Situation 2005. New York: United Nations.
United Nations (2005b). The Millennium Development Goals Report 2005. New
York: United Nations.
United Nations (2005c, July). Report of the Fact-Finding Mission to Zimbabwe World Bank (2000). Cities in Transition. Washington, D.C.
to Assess the Scope and Impact of Operation Murambatsyina by the UN
Special Envoy on Human Settlements Issues in Zimbabwe, Mrs. Anna Kajumulo World Bank (2000/2001). World Development Report 2000/2001: Attacking
Tibaijuka. Nairobi. Retrieved from http://www.unhabitat.org. Poverty. Washington, D.C.: World Bank.
United Nations Department of Economic and Social Affairs, Population Division World Bank (2003, June). Land Policies for Growth and Poverty Reduction.
(2004). World Urbanization Prospects: The 2003 Revision. New York. Retrieved Washington, D.C.: World Bank.
September 2005 from http://esa.un.org/unup.
World Bank (2005). World Development Report 2006. Washington, D.C.: Oxford
United Nations Department of Economic and Social Affairs, Population Division University Press.
(2005). World Population Prospects: The 2004 Revision. New York.
World Health Organization (1999). “Creating healthy cities in the 21st century”.
United Nations Department of Economic and Social Affairs, Statistics Division In: D. Satterthwaite (Ed.), The Earthscan Reader in Sustainable Cities. London:
(2005). Progress Towards the Millennium Development Goals, 1990-2005: Earthscan.
Goal 7 – Ensure Environmental Sustainability. Retrieved 31 August 2005 from
http://unstats.un.org/unsd/mi/goals_2005/Goal_7_2005.pdf. World Health Organization (2002). World Health Report 2002. Geneva: WHO.
United Nations General Assembly Economic and Social Council (2004, 6 World Health Organization. (2004). World Report on Road Traffic Injury
December). World Youth Report 2005: Report of the Secretary-General. Prevention. Geneva: WHO.
A/60/61-E/2005/7. Retrieved 20 January 2006 from www.un.org/youth.
Worldwatch Institute (2005). State of the World 2005: Redefining Global
United Nations Office for the Coordination of Humanitarian Affairs (OCHA) Security. New York and London: W.W. Norton & Company.
(2005a). Disaster Reduction and the Human Cost of Disaster: IRIN Web Special.
Retrieved June 2005 from http://www.irinnews.org. Worldwatch Institute (2006a, 24 January). “Rising toll from disasters
underscores need for humanitarian, political action”. [Media Release].
United Nations Office for the Coordination of Humanitarian Affairs (OCHA)
(2005b). IRIN News. Retrieved December 2005 from http://www.irinnews.org. Worldwatch Institute (2006b). State of the World 2006: Special Focus on India
and China. New York: W.W. Norton & Co.
US Census Bureau (2000). “Population profile of the United States: 2000”.
[Media Release]. Retrieved December 2005 from
http://www.census.gov/population/pop-profile/2000/chap07.pdf. Y
US Conference of Mayors (2004). US Metro Economies: A Decade of Prosperity. YUVA & Montgomery Watson Consultants, India (2000-2001). Final Report for
Retrieved December 2005 from http://www.usmayors.org/uscm. Slum Sanitation Undertaken for Mumbai Sewerage Disposal Project (MSDP).
Mumbai.
US Department of Housing and Urban Development, Office of Policy
Development and Research (2005). Assessing Housing Durability: A Pilot Study.
Washington, D.C. Z
US Library of Congress, Federal Research Division (1998). Sri Lanka: A Country Zulu, E. M., Dodoo, N. A, & Ezeh, A. C. (2002). “Sexual risk-taking in the slums of
Study. Washington, D.C. Nairobi, Kenya 1993-98”. Population Studies, 56, 311-332.
V
Vanderschueren, F. (2000, May). The Prevention of Urban Crime. Paper presented
at the Africities 2000 Summit. Windhoek, Namibia.
Vanderschueren, F. & Vezina, C. (2003). Reviewing Safer Cities. Background
Paper presented at the International Conference on Sustainable Safety:
Municipalities at the Crossroads, 25-28 November 2003, Durban, South Africa.
W
Walsh, B., Lewis, K.R. & McQuaid, J. (2005, 1 September). “Rebuilding New
Orleans: There may be a will, but is there a way?” Newhouse News Service.
Warah, R. (2002). “If you want to mobilize people, go to the public toilets”.
Habitat Debate, 8(2).
Watkins, K. (2000). The Oxfam Education Report. Oxford: Oxfam.
Watts, J. (2005, 31 October). “Satellite data reveals Beijing as pollution capital
of the world”. The Guardian.
WHO (2002). World Report on Violence and Health. Red Posters from the Global
Campaign for Violence Prevention. Geneva: WHO.
WHO (2005). The Protocol on Water and Health: Making a Difference. WHO
Regional Office for Europe.
WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation (2000).
Global Water Supply and Sanitation Assessment 2000 Report. Geneva: WHO.
WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation
(2004). Meeting the MDG Drinking Water and Sanitation Target: A Mid-Term
Assessment of Progress. Geneva. Retrieved September 2005 from
https://www.unicef.org.uk/publications.
Wilkinson, R. & Marmot, M. (Eds.) (2003). Social Determinants of Health. The
Solid Facts. (2nd ed.) Geneva: World Health Organization.
187
S TAT E O F T H E W O R L D ’ S C I T I E S R E P O R T 2 0 0 6 / 7
Indicators for the Habitat Agenda
and the Millennium Development Goals Indicators MDGs
1. Shelter
Promote the right to adequate housing Key indicator 1: durable structures Goal 7, targets 11, 32
Key indicator 2: overcrowding Goal 7, targets 11, 32
checklist 1: right to adequate housing
extensive indicator 1: housing price and rent-to-income
Provide security of tenure Key indicator 3: secure tenure Goal 7, targets11, 32
extensive indicator 2: authorized housing
extensive indicator 3: evictions
Provide equal access to credit checklist 2: housing finance

Statistical Annex Provide equal access to land

Promote access to basic services
extensive indicator 4: land price-to-income
Key indicator 4: access to safe water Goal 7, targets 10,30
Key indicator 5: access to improved sanitation Goal 7, targets 10,31
Monitoring the Habitat Agenda and the Millennium Development Goals Key indicator 6: connection to services

2. Social development and eradication of poverty

Provide equal opportunities for a safe and healthy life Key indicator 7: under-five mortality Goal 4, targets 5,13
Key indicator 8: homicides
checklist 3: urban violence
General disclaimer
extensive indicator 5: HIV prevalence Goal 6, targets 7, 18

The designations employed and presentation of the data in the Promote social integration and support disadvantaged groups Key indicator 9: poor households Goal 1, targets 11, 1
Statistical Annex do not imply the expression of any opinion what- Promote gender equality in human settlements development Key indicator 10: literacy rates Goal 3, targets 4, 10
soever on the part of the Secretariat of the United Nations concern- checklist 4: gender inclusion
extensive indicator 6: school enrolment Goal 3, targets 4, 10
ing the legal status of any country, city or area or of its authorities, extensive indicator 7: women councillors Goal 3, targets 4, 12
or concerning the delimitation of its frontiers or boundaries.
3. Environmental Management
Promote geographically-balanced settlement structures Key indicator 11: urban population growth
Key indicator 12: planned settlements
Manage supply and demand for water in an effective manner Key indicator 13: price of water
extensive indicator 8: water consumption

Reduce urban pollution Key indicator 14: wastewater treated

Prevent disasters and rebuild settlements
Promote effective and environmentally sound transportation systems
Support mechanisms to prepare and implement local environmental plans and
local Agenda 21 initiatives
Key indicator 15: solid waste disposal
extensive indicator 9: regular solid waste collection
checklist 5: disaster prevention and mitigation instruments
extensive indicator 10: houses in hazardous locations
Key indicator 16: travel time
extensive indicators 11: transport modes
Checklist 6: local environmental plans

4. Economic Development
Strengthen small and micro-enterprises, particularly those developed by women Key indicator 17: informal employment

Encourage public-private sector partnership and stimulate productive employ- Key indicator 18: city product
ment opportunities Key indicator 19: unemployment

5. Governance
Promote decentralisation and strengthen local authorities Key indicator 20: local government revenue
Checklist 7: decentralization Goal 8, targets 16, 45
Encourage and support participation and civic engagement Checklist 8: citizen participation
extensive indicator 12: voter participation
extensive indicator 13: civic associations
Ensure transparent, accountable and efficient governance of towns, cities and Checklist 9: transparency and accountability
metropolitan areas
Not Habitat Agenda but MDGs indicators Child malnutrition Goal 1, targets 2, 4
Immunization against measles Goal 4, targets 5,15
Births attended by skilled health personnel Goal 5, targets 6, 17
Solid fuel Goal 7, targets 9, 29
Not Habitat Agenda nor MDGs indicators Prevalence of diarrhoea and prevalence of Acute respiratory
infections (ARI)
188 189
TABLE 1: POPULATION OF SLUM AREAS AT MID-YEAR, BY REGION AND COUNTRY; 1990, 2001 AND SLUM ANNUAL GROWTH RATE

1990 2001 1990 2001

Total Urban Slum Total Urban Slum Slum Total Urban Slum Total Urban Slum Slum
Population Population Percentage Percentage Population Population Population Percentage Percentage Population Annual Growth Population Population Percentage Percentage Population Population Population Percentage Percentage Population Annual Growth
(thousands) (thousands) Urban Slum (thousands) (thousands) (thousands) Urban Slum (thousands) Rate (%) (thousands) (thousands) Urban Slum (thousands) (thousands) (thousands) Urban Slum (thousands) Rate (%)

MDG regional classification Belize 186 89 47.8 54.2 48 231 111 48.1 62.0 69 3.23
WORLD 5,254,807 2,285,693 43.5 31.3 714,972 6,134,124 2,923,184 47.7 31.2 912,918 2.22 Bolivia 6,573 3,653 55.6 70.0 2,555 8,516 5,358 62.9 61.3 3,284 2.28
Developed regions 933,494 694,260 74.4 6.0 41,750 985,592 753,909 76.5 6.0 45,191 0.72 Brazil 147,957 110,610 74.8 45.0 49,806 172,559 141,041 81.7 36.6 51,676 0.34
EURASIA (Countries in CIS) 281,610 184,261 65.4 10.3 18,929 282,639 181,182 64.1 10.3 18,714 -0.10 British Virgin Islands 17 9 52.9 3.0 0 24 15 62.0 3.0 0 4.64
European countries in CIS 214,807 152,222 70.9 6.0 9,208 208,208 147,673 70.9 6.0 8,878 -0.33 Cayman Islands 26 26 100.0 2.0 1 40 40 100.0 2.0 1 3.92
Asian countries in CIS 66,803 32,039 48.0 30.3 9,721 74,431 33,509 45.0 29.4 9,836 0.11 Chile 13,100 10,908 83.3 4.0 432 15,402 13,254 86.1 8.6 1,143 8.85
Developing regions 4,039,703 1,407,172 34.8 46.5 654,294 4,865,893 1,988,093 40.9 42.7 849,013 2.37 Colombia 34,970 24,029 68.7 26.0 6,239 42,803 32,319 75.5 21.8 7,057 1.12
Northern Africa 118,347 57,602 48.7 37.7 21,719 145,581 75,693 52.0 28.2 21,355 -0.15 Costa Rica 3,049 1,637 53.7 11.9 195 4,112 2,448 59.5 12.8 313 4.31
Sub-Saharan Africa 501,133 139,644 27.9 72.3 100,973 667,022 231,052 34.6 71.9 166,208 4.53 Cuba 10,629 7,828 73.6 2.0 156 11,237 8,482 75.5 2.0 169 0.73
Latin America and the Caribbean 440,419 312,995 71.1 35.4 110,837 526,594 399,322 75.8 31.9 127,566 1.28 Dominica 71 48 67.6 16.6 8 71 50 71.4 14.0 7 -1.17
Eastern Asia 1,226,423 367,210 29.9 41.1 150,761 1,364,438 533,182 39.1 36.4 193,824 2.28 Dominican Republic 7,061 4,126 58.4 56.4 2,327 8,507 5,615 66.0 37.6 2,111 -0.88
Eastern Asia excluding China (optional) 71,118 50,641 71.2 25.3 12,831 79,466 61,255 77.1 25.4 15,568 1.76 Ecuador 10,264 5,654 55.1 28.1 1,588 12,880 8,171 63.4 25.6 2,095 2.52
South Asia 1,173,908 311,867 26.6 63.7 198,663 1,449,417 428,677 29.6 59.0 253,122 2.20 El Salvador 5,112 2,517 49.2 44.7 1,126 6,400 3,935 61.5 35.2 1,386 1.89
South-eastern Asia 440,461 133,195 30.2 36.8 48,986 529,764 202,854 38.3 28.0 56,781 1.34 Falkland Is (Malvinas) 2 2 100.0 2.0 0 2 2 81.3 2.0 0 0.00
Western Asia 132,946 83,229 62.6 26.4 22,006 175,322 115,241 65.7 25.7 29,658 2.71 French Guiana 116 87 75.0 12.9 11 170 128 75.2 12.9 16 3.51
Oceania 6,066 1,430 23.6 24.5 350 7,755 2,072 26.7 24.1 499 3.24 Greenland 56 44 78.6 18.5 8 56 46 82.3 18.5 9 0.40
Grenada 91 31 34.1 6.9 2 94 36 38.4 6.9 2 1.36
Optional grouping Guadeloupe 391 385 98.5 6.9 27 431 430 99.6 6.9 30 1.00
Landlocked Developing Countries (LLDCs) 297,396 96,106 32.3 48.4 46,509 275,262 83,708 30.4 56.5 47,303 0.15 Guatemala 8,749 3,333 38.1 65.8 2,192 11,687 4,668 39.9 61.8 2,884 2.49
Small Island Developing States (SIDS) 44,908 23,852 53.1 24.0 5,735 52,644 30,083 57.1 24.4 7,327 2.23 Guyana 731 243 33.2 4.9 12 763 280 36.7 4.9 14 1.29
Least Developed Countries (LDCs) 515,348 107,341 20.8 76.3 81,925 685,365 179,295 26.2 78.2 140,121 4.88 Haiti 6,907 2,035 29.5 84.9 1,728 8,270 3,004 36.3 85.7 2,574 3.63
Honduras 4,870 2,036 41.8 24.0 488 6,575 3,531 53.7 18.1 638 2.43
List of countries Jamaica 2,369 1,219 51.5 29.2 356 2,598 1,470 56.6 35.7 525 3.53
EURASIA (Countries in CIS) 281,610 184,261 65.4 10.3 18,929 282,639 181,182 64.1 10.3 18,714 -0.10 Martinique 360 326 90.6 2.0 6 386 367 95.2 2.0 7 1.08
Mexico 83,223 60,303 72.5 23.1 13,923 100,368 74,846 74.6 19.6 14,692 0.49
Developing regions Montserrat 11 1 9.1 10.7 0 3 - 13.1 8.8 -
Northern Africa 118,347 57,602 48.7 37.7 21,719 145,581 75,693 52.0 28.2 21,355 -0.15 Netherlands Antilles 188 128 68.1 1.0 1 217 151 69.3 1.0 2 1.50
Algeria 24,855 12,776 51.4 11.8 1,508 30,841 17,801 57.7 11.8 2,101 3.02 Nicaragua 3,824 2,029 53.1 80.7 1,638 5,208 2,943 56.5 80.9 2,382 3.41
Egypt 56,223 24,499 43.6 57.5 14,087 69,080 29,475 42.7 39.9 11,762 -1.64 Panama 2,398 1,288 53.7 30.8 397 2,899 1,639 56.5 30.8 505 2.19
Libyan Arab Jamahiriya 4,311 3,528 81.8 35.2 1,242 5,408 4,757 88.0 35.2 1,674 2.72 Paraguay 4,219 2,054 48.7 36.8 756 5,636 3,194 56.7 25.0 797 0.48
Morocco 24,624 11,917 48.4 37.4 4,457 30,430 17,082 56.1 32.7 5,579 2.04 Peru 21,569 14,862 68.9 60.4 8,979 26,093 19,084 73.1 68.1 12,993 3.36
Tunisia 8,156 4,726 57.9 9.0 425 9,562 6,329 66.2 3.7 234 -5.43 Puerto Rico 3,528 2,516 71.3 2.0 50 3,952 2,987 75.6 2.0 59 1.56
Western Sahara 178 156 87.6 - - 260 249 95.7 2.0 5 Saint Kitts and Nevis 42 14 33.3 5.0 1 38 13 34.2 5.0 1 -0.67
Saint Lucia 131 49 37.4 11.9 6 149 57 38.0 11.9 7 1.37
Sub-Saharan Africa 501,133 139,644 27.9 72.3 100,973 667,022 231,052 34.6 71.9 166,208 4.53 St Vincent & the Grenadines 106 43 40.6 5.0 2 114 64 56.0 5.0 3 3.62
Angola 9,570 2,639 27.6 83.1 2,193 13,527 4,715 34.9 83.1 3,918 5.28 Saint-Pierre-et-Miquelon 6 6 100.0 8.7 1 7 6 92.2 8.7 1
Benin 4,655 1,605 34.5 80.3 1,288 6,446 2,774 43.0 83.6 2,318 5.34 Suriname 402 263 65.4 6.9 18 419 313 74.8 6.9 22 1.58
Botswana 1,240 525 42.3 59.2 311 1,554 768 49.4 60.7 466 3.69 Trinidad and Tobago 1,215 840 69.1 34.7 292 1,300 969 74.5 32.0 310 0.55
Burkina Faso 9,008 1,221 13.6 80.9 987 11,856 1,999 16.9 76.5 1,528 3.97 Turks and Caicos Islands 12 5 41.7 2.0 0 17 8 45.6 2.0 0 4.27
Burundi 5,636 353 6.3 83.3 294 6,502 603 9.3 65.3 394 2.66 Uruguay 3,106 2,763 89.0 6.9 191 3,361 3,097 92.1 2.0 62 -10.27
Cameroon 11,614 4,679 40.3 62.1 2,906 15,203 7,558 49.7 67.0 5,064 5.05 US Virgin Islands 104 46 44.2 2.0 1 122 57 46.7 6.9 4 13.25
Cape Verde 341 151 44.3 70.3 106 437 277 63.5 69.6 193 5.42 Venezuela 19,502 16,378 84.0 40.7 6,664 24,632 21,475 87.2 40.7 8,738 2.46
Central African Rep 2,945 1,104 37.5 94.0 1,038 3,782 1,575 41.7 92.4 1,455 3.07
Chad 5,829 1,227 21.0 99.3 1,218 8,135 1,964 24.1 99.1 1,947 4.26 Eastern Asia 1,226,423 367,210 29.9 41.1 150,761 1,364,438 533,182 39.1 36.4 193,824 2.28
Comoros 527 147 27.9 61.7 91 727 246 33.8 61.2 151 4.61 China 1,155,305 316,569 27.4 43.6 137,929 1,284,972 471,927 36.7 37.8 178,256 2.33
Congo 2,230 1,243 55.7 84.5 1,050 3,110 2,056 66.1 90.1 1,852 5.15 Hong Kong SAR of China 5,705 5,701 99.9 2.0 113 6,961 6,961 100.0 2.0 139 1.82
Côte d'Ivoire 12,582 5,014 39.9 50.5 2,532 16,349 7,197 44.0 67.9 4,884 5.97 Macao SAR of China 372 367 98.7 2.0 7 449 444 98.9 2.0 9 1.73
Dem Rep of the Congo 36,999 10,340 27.9 51.9 5,366 52,522 16,120 30.7 49.5 7,985 3.61 Korea, Dem People's Rep of 19,956 11,651 58.4 1.0 117 22,428 13,571 60.5 0.7 95 -1.86
Djibouti 504 408 81.0 - - 644 542 84.2 Korea, Rep of 42,869 31,658 73.8 37.0 11,728 47,069 38,830 82.5 37.0 14,385 25.55
Equatorial Guinea 352 126 35.8 89.1 112 470 232 49.3 86.5 201 5.28 Mongolia 2,216 1,264 57.0 68.5 866 2,559 1,449 56.6 64.9 940 0.75
Eritrea 3,103 490 15.8 69.9 342 3,816 730 19.1 69.9 510 3.62
Ethiopia 47,509 6,044 12.7 99.0 5,984 64,459 10,222 15.9 99.4 10,159 4.81 South Asia 1,173,908 311,867 26.6 63.7 198,663 1,449,417 428,677 29.6 59.0 253,122 2.20
Gabon 935 637 68.1 56.1 357 1,262 1,038 82.3 66.2 688 5.95 Afghanistan 13,675 2,495 18.2 98.5 2,458 22,474 5,019 22.3 98.5 4,945 6.35
Gambia 928 231 24.9 67.0 155 1,337 418 31.3 67.0 280 5.39 Bangladesh 110,025 21,750 19.8 87.3 18,988 140,369 35,896 25.6 84.7 30,403 4.28
Ghana 15,138 5,078 33.5 80.4 4,083 19,734 7,177 36.4 69.6 4,993 1.83 Bhutan 1,696 87 5.1 70.0 61 2,141 158 7.4 44.1 70 1.22
Guinea 6,139 1,439 23.4 79.6 1,145 8,274 2,312 27.9 72.3 1,672 3.44 India 844,886 215,747 25.5 60.8 131,174 1,025,096 285,608 27.9 55.5 158,418 1.72
Guinea-Bissau 946 225 23.8 93.4 210 1,227 397 32.3 93.4 371 5.17 Iran (Islamic Republic of) 58,435 32,917 56.3 51.9 17,094 71,369 46,204 64.7 44.2 20,406 1.61
Kenya 23,574 5,660 24.0 70.4 3,985 31,293 10,751 34.4 70.7 7,605 5.88 Maldives 216 56 25.9 0.0 - 300 84 28.0 0.0 -
Lesotho 1,682 338 20.1 49.8 168 2,057 592 28.8 57.0 337 6.32 Nepal 18,142 1,624 9.0 96.9 1,574 23,593 2,874 12.2 92.4 2,656 4.76
Liberia 2,144 900 42.0 70.2 632 3,108 1,414 45.5 55.7 788 2.00 Pakistan 109,811 33,565 30.6 78.7 26,416 144,971 48,425 33.4 73.6 35,627 2.72
Madagascar 11,956 2,818 23.6 90.9 2,562 16,437 4,952 30.1 92.9 4,603 5.33 Sri Lanka 17,022 3,626 21.3 24.8 899 19,104 4,409 23.1 13.6 597 -3.72
Malawi 9,434 1,092 11.6 94.6 1,033 11,572 1,745 15.1 91.1 1,590 3.92
Mali 8,778 2,091 23.8 94.1 1,968 11,677 3,606 30.9 93.2 3,361 4.87 South-eastern Asia 440,461 133,195 30.2 36.8 48,986 529,764 202,854 38.3 28.0 56,781 1.34
Mauritania 1,992 877 44.0 94.3 827 2,747 1,624 59.1 94.3 1,531 5.60 Brunei Darussalam 257 169 65.8 2.0 3 335 244 72.8 2.0 5 3.34
Mauritius 1,057 428 40.5 - 1,171 486 41.6 - - Cambodia 9,630 1,213 12.6 71.7 870 13,441 2,348 17.5 72.2 1,696 6.07
Mozambique 13,645 2,880 21.1 94.5 2,722 18,644 6,208 33.3 94.1 5,841 6.94 Indonesia 182,474 55,819 30.6 32.2 17,964 214,840 90,356 42.1 23.1 20,877 1.37
Namibia 1,375 366 26.6 42.3 155 1,788 561 31.4 37.9 213 2.88 Lao People's Dem Republic 4,132 638 15.4 66.1 422 5,403 1,066 19.7 66.1 705 4.67
Niger 7,707 1,241 16.1 96.0 1,191 11,227 2,366 21.1 96.2 2,277 5.89 Malaysia 17,845 8,891 49.8 2.0 177 22,633 13,154 58.1 2.0 262 3.56
Nigeria 85,953 30,120 35.0 80.0 24,096 116,929 52,539 44.9 79.2 41,595 4.96 Myanmar 40,517 9,984 24.6 31.1 3,105 48,364 13,606 28.1 26.4 3,596 1.34
Réunion 604 386 63.9 - - 732 528 72.1 Philippines 61,040 29,774 48.8 54.9 16,346 77,131 45,812 59.4 44.1 20,183 1.92
Rwanda 6,766 360 5.3 82.2 296 7,949 497 6.3 87.9 437 3.55 Singapore 3,016 3,016 100.0 0.0 - 4,108 4,108 100.0 0.0 -
Saint Helena 6 3 50.0 - - 6 5 71.9 2.0 0 Thailand 54,736 10,244 18.7 19.5 1,998 63,584 12,709 20.0 2.0 253 -18.79
Sao Tome & Principe 115 45 39.1 - - 140 67 47.7 2.0 1 Timor-Leste 740 58 7.8 2.0 1 750 56 7.5 12.0 7 16.00
Senegal 7,327 2,933 40.0 77.6 2,276 9,662 4,653 48.2 76.4 3,555 4.05 Viet Nam 66,074 13,389 20.3 60.5 8,100 79,175 19,395 24.5 47.4 9,197 1.15
Seychelles 70 37 52.9 - - 81 53 64.6 2.0 1
Sierra Leone 4,061 1,218 30.0 90.9 1,107 4,587 1,714 37.3 95.8 1,642 3.58 Western Asia 132,946 83,229 62.6 26.4 22,006 175,322 115,241 65.7 25.7 29,658 2.71
Somalia 7,163 1,734 24.2 96.3 1,670 9,157 2,557 27.9 97.1 2,482 3.60 Bahrain 490 429 87.6 0.0 - 652 603 92.5 2.0 12
South Africa 36,376 17,763 48.8 46.2 8,207 43,792 25,260 57.7 33.2 8,376 0.19 Cyprus 681 442 64.9 0.0 - 790 555 70.2 0.0 -
Sudan 24,818 6,606 26.6 86.4 5,708 31,809 11,790 37.1 85.7 10,107 5.19 Iraq 17,271 12,027 69.6 56.7 6,825 23,584 15,907 67.4 56.7 9,026 2.54
Swaziland 769 183 23.8 - - 938 250 26.7 Israel 4,514 4,074 90.3 2.0 81 6,172 5,666 91.8 2.0 113 3.00
Togo 3,453 984 28.5 80.9 796 4,657 1,579 33.9 80.6 1,273 4.27 Jordan 3,254 2,350 72.2 16.5 388 5,051 3,979 78.7 15.7 623 4.32
Uganda 17,245 1,925 11.2 93.8 1,806 24,023 3,486 14.5 93.0 3,241 5.32 Kuwait 2,143 2,034 94.9 3.0 60 1,971 1,894 96.1 3.0 56 -0.65
U. Rep of Tanzania 26,043 5,652 21.7 99.1 5,601 35,965 11,982 33.3 92.1 11,031 6.16 Lebanon 2,713 2,284 84.2 50.0 1,142 3,556 3,203 90.1 50.0 1,602 3.07
Zambia 8,049 3,172 39.4 72.0 2,284 10,649 4,237 39.8 74.0 3,136 2.88 Occupied Palestinian Territory 2,154 1,379 64.0 - - 3,311 2,222 67.1 60.0 1,333
Zimbabwe 10,241 2,906 28.4 4.0 116 12,852 4,630 36.0 3.4 157 2.76 Oman 1,785 1,109 62.1 60.5 671 2,622 2,006 76.5 60.5 1,214 5.39
Qatar 453 407 89.8 2.0 8 575 534 92.9 2.0 11 2.47
Latin America and the Caribbean 440,419 312,995 71.1 35.4 110,837 526,594 399,322 75.8 31.9 127,566 1.28 Saudi Arabia 15,400 12,046 78.2 19.8 2,385 21,028 18,229 86.7 19.8 3,609 3.77
Anguilla 8 8 100.0 40.6 3 12 12 100.0 40.6 5 3.69 Syrian Arab Republic 12,386 6,061 48.9 10.4 629 16,610 8,596 51.8 10.4 892 3.18
Antigua and Barbuda 63 22 34.9 6.9 2 65 24 37.1 6.9 2 0.79 Turkey 56,098 34,324 61.2 23.3 7,997 67,632 44,755 66.2 17.9 8,011 0.02
Argentina 32,527 28,141 86.5 30.5 8,597 37,488 33,119 88.3 33.1 10,964 2.21 United Arab Emirates 2,014 1,615 80.2 2.0 32 2,654 2,314 87.2 2.0 46 3.27
Aruba 66 33 50.0 2.0 1 104 53 51.0 2.0 1 4.31 Yemen 11,590 2,648 22.8 67.5 1,787 19,114 4,778 25.0 65.1 3,110 5.03
Bahamas 255 213 83.5 2.0 4 308 274 88.9 2.0 5 2.29
Barbados 257 115 44.7 1.0 1 268 136 50.5 1.0 1 1.52

190 191
TABLE 2 : SLUM POPULATION PROJECTIONS, 1990-2020

SLUM POPULATION (THOUSAND) SLUM PROJECTION TAGET 11 (THOUSAND) SCENARIOS 2020 SLUM POPULATION (THOUSAND) SLUM PROJECTION TAGET 11 (THOUSAND) SCENARIOS
Moderate Reduce % Moderate Reduce %
No Change by half No Change by half
1990 2001 2005 2010 2015 2020 1990 2001 2005 2010 2015 2020 Change 100 mill. 1990-2020 1990 2001 2005 2010 2015 2020 1990 2001 2005 2010 2015 2020 Change 100 mill. 1990-2020

WORLD 714,972 912,918 997,767 1,115,002 1,246,012 1,392,416 714,972 912,918 976,858 1,070,494 1,175,132 1,292,065 1,392,416 1,292,065 705,745 Cayman Islands 1 1 1 1 1 2 1 1 1 1 1 2 2 2 1
Developed regions 41,750 45,191 46,511 48,216 49,983 51,815 41,750 45,191 45,507 46,167 46,851 47,560 51,815 47,560 26,137 Chile 432 1,143 1,628 2,534 3,943 6,136 432 1,143 1,598 2,456 3,791 5,868 6,136 5,868 3,110
EURASIA (Countries in CIS) 18,929 18,714 18,637 18,541 18,445 18,350 18,929 18,714 18,228 17,725 17,225 16,727 18,350 16,727 9,039 Colombia 6,239 7,057 7,381 7,806 8,256 8,732 6,239 7,057 7,223 7,480 7,752 8,039 8,732 8,039 4,426
European countries in CIS 9,208 8,878 8,761 8,617 8,475 8,336 9,208 8,878 8,568 8,234 7,906 7,583 8,336 7,583 Costa Rica 195 313 372 461 572 710 195 313 364 444 544 667 710 667 360
Asian countries in CIS 9,721 9,836 9,879 9,932 9,986 10,040 9,721 9,836 9,663 9,499 9,334 9,168 10,040 9,168 Cuba 156 169 174 180 187 194 156 169 170 173 175 178 194 178 98
Developing regions 654,294 849,013 933,376 1,050,714 1,182,803 1,331,498 654,294 849,013 913,874 1,009,026 1,116,140 1,236,719 1,331,498 1,236,719 670,570 Dominica 8 7 7 6 6 6 8 7 7 6 6 5 6 5 3
Northern Africa 21,719 21,355 21,224 21,062 20,901 20,741 21,719 21,355 20,758 20,133 19,513 18,898 20,741 18,898 17,286 Dominican Republic 2,327 2,111 2,038 1,950 1,865 1,785 2,327 2,111 1,992 1,861 1,735 1,615 1,785 1,615 905
Sub-Saharan Africa 100,973 166,208 199,231 249,886 313,419 393,105 100,973 166,208 195,245 240,808 297,955 369,631 393,105 369,631 150,654 Ecuador 1,588 2,095 2,317 2,629 2,982 3,382 1,588 2,095 2,269 2,525 2,815 3,144 3,382 3,144 1,714
Latin America and the Caribbean 110,837 127,566 134,257 143,116 152,559 162,626 110,837 127,566 131,390 137,174 143,340 149,913 162,626 149,913 81,385 El Salvador 1,126 1,386 1,495 1,644 1,807 1,986 1,126 1,386 1,464 1,577 1,702 1,839 1,986 1,839 1,006
Eastern Asia 150,761 193,824 212,368 238,061 266,863 299,150 150,761 193,824 207,923 228,583 251,742 277,704 299,150 277,704 157,527 French Guiana 11 16 19 23 27 32 11 16 19 22 26 30 32 30 16
Eastern Asia excluding China (optional) 12,831 15,568 16,702 18,236 19,911 21,739 12,831 15,568 16,348 17,494 18,744 20,109 21,739 20,109 Greenland 8 9 9 9 9 9 8 9 8 8 8 8 9 8 5
South Asia 198,663 253,122 276,432 308,611 344,537 384,644 198,663 253,122 270,637 296,283 324,914 356,877 384,644 356,877 178,762 Grenada 2 2 3 3 3 3 2 2 3 3 3 3 3 3 2
South-eastern Asia 48,986 56,781 59,913 64,073 68,521 73,279 48,986 56,781 58,636 61,420 64,398 67,583 73,279 67,583 58,302 Guadeloupe 27 30 31 33 34 36 27 30 30 31 32 33 36 33 18
Western Asia 22,006 29,658 33,057 37,860 43,360 49,659 22,006 29,658 32,371 36,379 40,968 46,224 49,659 46,224 26,290 Guatemala 2,192 2,884 3,186 3,609 4,089 4,632 2,192 2,884 3,120 3,467 3,860 4,305 4,632 4,305 2,348
Oceania 350 499 568 668 786 924 350 499 557 643 744 863 924 863 363 Guyana 12 14 15 16 17 18 12 14 14 15 16 16 18 16 9
Haiti 1,728 2,574 2,976 3,568 4,277 5,128 1,728 2,574 2,916 3,434 4,054 4,799 5,128 4,799 2,599
Developing regions Honduras 488 638 703 793 896 1,012 488 638 688 762 846 940 1,012 940 513
Northern Africa 21,719 21,355 21,224 21,062 20,901 20,741 21,719 21,355 20,758 20,133 19,513 18,898 20,741 18,898 10,513 Jamaica 356 525 604 721 860 1,026 356 525 592 693 815 959 1,026 959 520
Algeria 1,508 2,101 2,370 2,755 3,204 3,725 1,508 2,101 2,321 2,649 3,030 3,474 3,725 3,474 1,888 Martinique 6 7 8 8 8 9 6 7 7 8 8 8 9 8 5
Egypt 14,087 11,762 11,015 10,148 9,349 8,613 14,087 11,762 10,766 9,671 8,662 7,733 8,613 7,733 4,365 Mexico 13,923 14,692 14,983 15,353 15,733 16,123 13,923 14,692 14,657 14,694 14,732 14,771 16,123 14,771 8,172
Libyan Arab Jamahiriya 1,242 1,674 1,867 2,138 2,450 2,806 1,242 1,674 1,828 2,055 2,314 2,612 2,806 2,612 1,422 Netherlands Antilles 1 2 2 2 2 2 1 2 2 2 2 2 2 2 1
Morocco 4,457 5,579 6,054 6,705 7,425 8,223 4,457 5,579 5,927 6,435 6,998 7,621 8,223 7,621 4,168 Nicaragua 1,638 2,382 2,730 3,237 3,837 4,550 1,638 2,382 2,674 3,114 3,635 4,253 4,550 4,253 2,306
Tunisia 425 234 188 144 110 84 425 234 184 136 99 71 84 71 42 Panama 397 505 552 615 687 766 397 505 540 591 648 711 766 711 388
Western Sahara - 5 - 5 Paraguay 756 797 812 832 852 873 756 797 795 796 798 800 873 800 443
Peru 8,979 12,993 14,862 17,581 20,796 24,601 8,979 12,993 14,558 16,911 19,695 22,988 24,601 22,988 12,469
Sub-Saharan Africa 100,973 166,208 199,231 249,886 313,419 393,105 100,973 166,208 195,245 240,808 297,955 369,631 393,105 369,631 199,245 Puerto Rico 50 59 63 68 74 80 50 59 62 66 70 74 80 74 41
Angola 2,193 3,918 4,839 6,300 8,201 10,677 2,193 3,918 4,743 6,077 7,814 10,075 10,677 10,075 5,412 Saint Kitts and Nevis 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0
Benin 1,288 2,318 2,870 3,749 4,896 6,394 1,288 2,318 2,814 3,617 4,666 6,035 6,394 6,035 3,241 Saint Lucia 6 7 7 8 8 9 6 7 7 7 8 8 9 8 4
Botswana 311 466 540 650 781 939 311 466 529 625 740 879 939 879 476 St Vincent & the Grenadines 2 3 4 4 5 6 2 3 4 4 5 6 6 6 3
Burkina Faso 987 1,528 1,791 2,185 2,665 3,250 987 1,528 1,755 2,104 2,529 3,047 3,250 3,047 1,647 Saint-Pierre-et-Miquelon 1 1 1 1
Burundi 294 394 438 501 572 653 294 394 429 481 540 608 653 608 331 Suriname 18 22 23 25 27 29 18 22 23 24 26 27 29 27 15
Cameroon 2,906 5,064 6,197 7,977 10,268 13,217 2,906 5,064 6,074 7,693 9,777 12,459 13,217 12,459 6,699 Trinidad and Tobago 292 310 317 326 335 344 292 310 310 312 314 315 344 315 174
Cape Verde 106 193 240 314 412 540 106 193 235 303 393 510 540 510 274 Uruguay 191 62 41 24 15 9 191 62 40 23 13 6 9 6 4
Central African Rep 1,038 1,455 1,646 1,919 2,238 2,610 1,038 1,455 1,612 1,845 2,117 2,435 2,610 2,435 1,323 US Virgin Islands 1 4 7 13 25 49 1 4 7 13 24 47 49 47 25
Chad 1,218 1,947 2,308 2,856 3,534 4,373 1,218 1,947 2,262 2,751 3,357 4,106 4,373 4,106 2,216 Venezuela 6,664 8,738 9,642 10,906 12,336 13,952 6,664 8,738 9,441 10,475 11,645 12,967 13,952 12,967 7,072
Comoros 91 151 181 228 287 361 91 151 177 220 273 340 361 340 183
Congo 1,050 1,852 2,276 2,945 3,810 4,930 1,050 1,852 2,231 2,840 3,629 4,650 4,930 4,650 2,499 Eastern Asia 150,761 193,824 212,368 238,061 266,863 299,150 150,761 193,824 207,923 228,583 251,742 277,704 299,150 277,704 151,624
Côte d'Ivoire 2,532 4,884 6,203 8,361 11,271 15,194 2,532 4,884 6,082 8,074 10,760 14,381 15,194 14,381 7,701 China 137,929 178,256 195,682 219,878 247,066 277,616 137,929 178,256 191,590 211,141 233,109 257,793 277,616 257,793 140,709
Dem Rep of the Congo 5,366 7,985 9,227 11,054 13,243 15,865 5,366 7,985 9,039 10,637 12,552 14,846 15,865 14,846 8,041 Hong Kong SAR of China 113 139 149 163 179 196 113 139 146 156 168 181 196 181 99
Equatorial Guinea 112 201 248 323 420 547 112 201 243 311 400 516 547 516 277 Macao SAR of China 7 9 9 10 11 12 7 9 9 10 11 11 12 11 6
Eritrea 342 510 590 707 847 1,016 342 510 578 680 803 950 1,016 950 515 Korea, Dem People's Rep of 117 95 88 80 73 67 117 95 86 77 68 60 67 60 34
Ethiopia 5,984 10,159 12,315 15,665 19,926 25,347 5,984 10,159 12,070 15,102 18,960 23,866 25,347 23,866 12,847 Korea, Rep of 11,728 14,385 15,494 17,002 18,655 20,470 11,728 14,385 15,167 16,313 17,569 18,948 20,470 18,948 10,779
Gabon 357 688 872 1,174 1,581 2,129 357 688 855 1,134 1,509 2,015 2,129 2,015 1,079 Mongolia 866 940 969 1,006 1,044 1,084 866 940 948 963 979 995 1,084 995 571
Gambia 155 280 348 455 596 781 155 280 341 439 568 737 781 737 396
Ghana 4,083 4,993 5,372 5,886 6,450 7,067 4,083 4,993 5,258 5,647 6,073 6,540 7,067 6,540 3,582 South Asia 198,663 253,122 276,432 308,611 344,537 384,644 198,663 253,122 270,637 296,283 324,914 356,877 384,644 356,877 194,957
Guinea 1,145 1,672 1,918 2,278 2,705 3,213 1,145 1,672 1,879 2,192 2,563 3,003 3,213 3,003 1,628 Afghanistan 2,458 4,945 6,375 8,760 12,036 16,536 2,458 4,945 6,252 8,464 11,502 15,676 16,536 15,676 8,381
Guinea-Bissau 210 371 456 591 765 990 210 371 447 570 728 934 990 934 502 Bangladesh 18,988 30,403 36,079 44,687 55,348 68,553 18,988 30,403 35,353 43,047 52,576 64,378 68,553 64,378 34,746
Kenya 3,985 7,605 9,620 12,905 17,311 23,223 3,985 7,605 9,432 12,460 16,522 21,972 23,223 21,972 11,771 Bhutan 61 70 73 78 83 88 61 70 72 75 78 81 88 81 45
Lesotho 168 337 434 596 817 1,121 168 337 426 576 781 1,062 1,121 1,062 568 India 131,174 158,418 169,671 184,868 201,425 219,466 131,174 158,418 166,079 177,332 189,592 202,950 219,466 202,950 111,236
Liberia 632 788 853 943 1,043 1,153 632 788 835 905 983 1,068 1,153 1,068 584 Iran (Islamic Republic of) 17,094 20,406 21,763 23,587 25,564 27,707 17,094 20,406 21,301 22,621 24,052 25,603 27,707 25,603 14,043
Madagascar 2,562 4,603 5,696 7,434 9,703 12,664 2,562 4,603 5,583 7,172 9,246 11,953 12,664 11,953 6,419 Nepal 1,574 2,656 3,213 4,077 5,172 6,562 1,574 2,656 3,149 3,930 4,920 6,177 6,562 6,177 3,326
Malawi 1,033 1,590 1,860 2,262 2,752 3,348 1,033 1,590 1,822 2,178 2,611 3,138 3,348 3,138 1,697 Pakistan 26,416 35,627 39,722 45,507 52,136 59,730 26,416 35,627 38,897 43,728 49,262 55,602 59,730 55,602 30,274
Mali 1,968 3,361 4,083 5,208 6,643 8,474 1,968 3,361 4,002 5,022 6,322 7,981 8,474 7,981 4,295 Sri Lanka 899 597 515 428 355 295 899 597 503 406 325 258 295 258 149
Mauritania 827 1,531 1,915 2,534 3,353 4,437 827 1,531 1,878 2,446 3,198 4,193 4,437 4,193 2,249
Mozambique 2,722 5,841 7,710 10,909 15,437 21,842 2,722 5,841 7,563 10,549 14,775 20,753 21,842 20,753 11,071 South-eastern Asia 48,986 56,781 59,913 64,073 68,521 73,279 48,986 56,781 58,636 61,420 64,398 67,583 73,279 67,583 37,141
Namibia 155 213 239 276 318 368 155 213 234 265 301 343 368 343 186 Brunei Darussalam 3 5 6 7 8 9 3 5 5 6 7 9 9 9 5
Niger 1,191 2,277 2,882 3,869 5,194 6,972 1,191 2,277 2,826 3,736 4,957 6,597 6,972 6,597 3,534 Cambodia 870 1,696 2,162 2,929 3,968 5,375 870 1,696 2,120 2,829 3,789 5,089 5,375 5,089 2,724
Nigeria 24,096 41,595 46,272 55,732 66,026 76,749 24,096 41,595 48,507 57,422 67,037 76,943 76,749 76,943 38,900 Indonesia 17,964 20,877 22,049 23,608 25,277 27,064 17,964 20,877 21,579 22,632 23,759 24,965 27,064 24,965 13,718
Rwanda 296 437 504 601 718 857 296 437 493 579 681 802 857 802 435 Lao People's Dem Republic 422 705 850 1,073 1,355 1,711 422 705 833 1,034 1,289 1,610 1,711 1,610 867
Sao Tome & Principe - 1 - 1 Malaysia 177 262 302 361 431 515 177 262 296 347 408 482 515 482 261
Senegal 2,276 3,555 4,181 5,120 6,270 7,679 2,276 3,555 4,096 4,930 5,952 7,203 7,679 7,203 3,892 Myanmar 3,105 3,596 3,794 4,056 4,336 4,635 3,105 3,596 3,713 3,888 4,075 4,275 4,635 4,275 2,349
Seychelles - 1 - 1 Philippines 16,346 20,183 21,792 23,984 26,397 29,053 16,346 20,183 21,333 23,015 24,866 26,904 29,053 26,904 14,725
Sierra Leone 1,107 1,642 1,895 2,266 2,711 3,243 1,107 1,642 1,856 2,181 2,569 3,034 3,243 3,034 1,644 Thailand 1,998 253 119 47 18 7 1,998 253 115 42 13 2 7 2 4
Somalia 1,670 2,482 2,867 3,433 4,111 4,923 1,670 2,482 2,809 3,304 3,896 4,606 4,923 4,606 2,495 Timor-Leste 1 7 13 28 63 140 1 7 13 28 61 136 140 136 71
South Africa 8,207 8,376 8,439 8,517 8,597 8,677 8,207 8,376 8,254 8,147 8,039 7,930 8,677 7,930 4,398 Viet Nam 8,100 9,197 9,632 10,204 10,811 11,453 8,100 9,197 9,426 9,779 10,152 10,548 11,453 10,548 5,805
Sudan 5,708 10,107 12,441 16,131 20,915 27,118 5,708 10,107 12,195 15,560 19,923 25,580 27,118 25,580 13,745
Togo 796 1,273 1,510 1,870 2,315 2,866 796 1,273 1,480 1,801 2,199 2,691 2,866 2,691 1,452 Western Asia 22,006 29,658 33,057 37,860 43,360 49,659 22,006 29,658 32,371 36,379 40,968 46,224 49,659 46,224 25,169
Uganda 1,806 3,241 4,010 5,231 6,825 8,904 1,806 3,241 3,931 5,047 6,503 8,403 8,904 8,403 4,513 Bahrain - 12 - 12
U. Rep of Tanzania 5,601 11,031 14,113 19,205 26,133 35,561 5,601 11,031 13,840 18,551 24,962 33,685 35,561 33,685 18,024 Iraq 6,825 9,026 9,992 11,346 12,884 14,630 6,825 9,026 9,784 10,899 12,166 13,604 14,630 13,604 7,415
Zambia 2,284 3,136 3,519 4,065 4,695 5,423 2,284 3,136 3,446 3,907 4,439 5,053 5,423 5,053 2,749 Israel 81 113 127 148 172 199 81 113 124 142 162 186 199 186 101
Zimbabwe 116 157 176 202 232 266 116 157 172 194 219 247 266 247 135 Jordan 388 623 741 920 1,141 1,416 388 623 726 886 1,084 1,330 1,416 1,330 718
Kuwait 60 56 55 53 51 50 60 56 54 51 48 45 50 45 25
Latin America and the Caribbean 110,837 127,566 134,257 143,116 152,559 162,626 110,837 127,566 131,390 137,174 143,340 149,913 162,626 149,913 82,427 Lebanon 1,142 1,602 1,811 2,112 2,463 2,872 1,142 1,602 1,774 2,031 2,330 2,679 2,872 2,679 1,456
Anguilla 3 5 6 7 8 10 3 5 6 7 8 9 10 9 5 Occupied Palestinian Territory - 1,333 - 1,333
Antigua and Barbuda 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 Oman 671 1,214 1,506 1,972 2,581 3,379 671 1,214 1,476 1,902 2,460 3,190 3,379 3,190 1,713
Argentina 8,597 10,964 11,978 13,379 14,943 16,690 8,597 10,964 11,727 12,844 14,092 15,486 16,690 15,486 8,459 Qatar 8 11 12 13 15 17 8 11 11 13 14 16 17 16 9
Aruba 1 1 1 2 2 2 1 1 1 1 2 2 2 2 1 Saudi Arabia 2,385 3,609 4,196 5,066 6,115 7,382 2,385 3,609 4,111 4,876 5,799 6,914 7,382 6,914 3,742
Bahamas 4 5 6 7 8 8 4 5 6 6 7 8 8 8 4 Syrian Arab Republic 629 892 1,012 1,187 1,391 1,630 629 892 992 1,141 1,316 1,522 1,630 1,522 826
Barbados 1 1 1 2 2 2 1 1 1 1 2 2 2 2 1 Turkey 7,997 8,011 8,016 8,022 8,029 8,035 7,997 8,011 7,841 7,671 7,501 7,332 8,035 7,332 4,072
Belize 48 69 78 92 108 127 48 69 77 88 102 119 127 119 64 United Arab Emirates 32 46 52 62 73 86 32 46 51 59 69 80 86 80 43
Bolivia 2,555 3,284 3,597 4,032 4,519 5,064 2,555 3,284 3,522 3,871 4,263 4,701 5,064 4,701 2,567 Yemen 1,787 3,110 3,803 4,892 6,292 8,092 1,787 3,110 3,728 4,717 5,990 7,628 8,092 7,628 4,102
Brazil 49,806 51,676 52,374 53,259 54,159 55,074 49,806 51,676 51,234 50,958 50,677 50,392 55,074 50,392 27,914
British Virgin Islands 0 0 1 1 1 1 0 0 1 1 1 1 1 1 1

192 193
TABLE 3 : POPULATION OF SLUM AREAS AT MID-YEAR BY SHELTER DEPRIVATION, BY REGION AND COUNTRY 1990 AND 2001 TABLE 4 : PROPORTION OF URBAN HOUSEHOLDS WITH FINISHED MAIN FLOOR MATERIALS, SUFFICIENT LIVING AREA, SUSTAINABLE ACCESS TO
SAFE WATER AND IMPROVED SANITATION

POPULATION OF SLUM AREAS AT MID-YEAR, 1990 POPULATION OF SLUM AREAS AT MID-YEAR, 2001
Finished main floor materials Access to sufficient living area Access to safe water source Access to improved sanitation
Percentage of slum by number of shelter deprivation Percentage of slum by number of shelter deprivation
Country City 1990 1993 1998 2000 2003 1990 1993 1998 2000 2003 1990 1993 1998 2000 2003 1990 1993 1998 2000 2003

One Two Three Four One Two Three Four

Percentage Percentage Shelter Shelter Shelter Shelter Percentage Percentage Shelter Shelter Shelter Shelter
Urban Slum Deprivation Deprivation Deprivation Deprivation Urban Slum Deprivation Deprivation Deprivation Deprivation Northern Africa 97.5 97.7 98.0 98.1 98.3 71.0 78.2 85.8 90.6 90.5 96.0 95.8 95.3 95.2 94.9 84.0 85.3 87.3 88.2 89.4
Sub-Saharan Africa 82.2 83.9 86.1 87.0 89.1 72.9 73.6 73.6 73.7 73.1 82.0 82.0 82.0 82.0 82.0 54.0 54.3 54.7 54.8 55.1
Latin America and the Caribbean 98.2 88.2 93.0 93.5 94.3 94.7 95.2 82.0 82.5 83.3 83.7 84.2
Northern Africa 48.7 37.7 30.1 4.9 2.2 0.5 52.0 28.2 25.0 3.0 0.2 -
Eastern Asia 98.4 91.5 99.0 97.5 95.0 94.0 92.5 64.0 65.3 67.3 68.2 69.4
Sub-Saharan Africa 27.9 72.3 32.1 27.0 10.8 2.4 34.6 71.9 35.5 24.0 10.6 1.9 Southern Asia 84.8 65.0 90.0 91.0 92.7 93.3 94.3 54.0 57.0 62.0 64.0 67.0
Latin America and the Caribbean 71.1 35.4 23.3 8.9 2.5 0.7 75.8 31.9 21.1 8.0 2.4 0.4 South-eastern Asia 91.9 92.3 93.0 93.3 93.6 66.0 67.9 71.0 73.2 73.1 91.0 91.0 91.0 91.0 91.0 67.0 70.0 75.0 77.0 80.0
Eastern Asia 29.9 41.1 39.1 36.4 Western Asia 96.7 96.8 96.7 96.8 96.4 91.3 92.2 93.5 94.2 91.1 94.0 94.3 94.7 94.8 95.1 96.0 95.8 95.3 95.2 94.9
South Asia 26.6 63.7 38.8 20.7 4.1 0.1 29.6 59.0 38.8 17.5 2.7 -
South-eastern Asia 30.2 36.8 17.8 12.0 5.4 1.6 38.3 28.0 20.7 5.7 1.4 0.2 Northern Africa
Western Asia 62.6 26.4 19.8 4.8 1.6 0.3 65.7 25.7 19.7 4.0 1.6 0.3 Algeria Oran 98.0 98.0 98.0 98.0 98.0
Algeria Constantine 98.0 98.0 98.0 98.0 98.0
Oceania 23.6 24.5 26.7 24.1
Algeria Blida 98.0 98.0 98.0 98.0 98.0
Algeria Sétif 98.0 98.0 98.0 98.0 92.0
Northern Africa
Algeria Tébessa 98.0 98.0 98.0 98.0 92.0
Egypt 43.6 57.5 45.6 7.0 4.0 0.9 42.7 39.9 35.4 4.5 0.1 - Algeria Wargla 98.0 98.0 98.0 98.0 92.0
Morocco 48.4 37.4 30.3 5.8 1.0 0.3 56.1 32.7 29.0 3.0 0.5 0.1 Algeria Midyah 98.0 98.0 98.0 98.0 92.0
Egypt Cairo 98.9 98.8 98.6 98.5 98.4 70.9 78.1 89.9 94.7 94.7 99.0 99.2 99.6 99.7 99.9 74.2 77.3 82.5 84.5 87.6
Sub-Saharan Africa Egypt Alexandria 97.7 97.7 97.9 97.9 98.0 70.4 77.7 89.9 94.7 94.7 99.6 99.7 99.7 99.7 99.8 79.5 82.6 87.7 89.7 92.8
Benin 34.5 80.3 36.3 24.1 15.9 3.9 43.0 83.6 46.6 24.1 12.9 - Egypt Port Said 97.2 97.6 98.4 98.6 99.1 75.5 81.8 92.4 96.7 96.7 97.7 97.5 97.1 97.0 96.7 89.1 90.7 93.3 94.4 96.0
Burkina Faso 13.6 80.9 37.6 28.6 13.7 1.1 16.9 76.5 51.1 22.9 2.1 0.4 Egypt Suez 99.4 99.0 98.2 97.9 97.4 72.9 80.2 92.5 97.4 97.4 99.4 99.5 99.5 99.5 99.6 81.4 83.4 86.6 87.9 89.8
Egypt Assyut 75.1 77.9 82.5 84.4 87.1 87.7 94.9 94.9 92.9 94.9 98.3 99.6 99.6 61.3 61.6 62.2 62.4 62.7
Cameroon 40.3 62.1 30.0 22.3 8.8 1.0 49.7 67.0 37.7 16.7 11.2 1.5
Egypt Aswan 70.7 74.1 79.8 82.1 85.5 85.8 93.6 93.6 97.2 97.8 98.8 99.1 99.7 62.3 63.8 66.2 67.1 68.6
Central African Rep 37.5 94.0 16.2 36.6 37.3 3.9 41.7 92.4 15.9 36.0 36.7 3.8
Egypt Beni Suef 52.2 64.3 84.5 92.6 92.6 85.6 92.3 92.3 87.3 90.6 95.9 98.1 98.1 47.7 55.0 67.1 72.0 79.3
Chad 21.0 99.3 10.3 25.6 45.9 17.5 24.1 99.1 10.3 25.5 45.9 17.5
Morocco Casablanca 98.7 99.0 99.4 99.6 99.9 69.2 74.7 82.8 100.0 100.0 99.9 99.8 99.7 71.5 77.4 87.1 91.1 96.9
Comoros 27.9 61.7 29.3 21.8 9.3 1.3 33.8 61.2 29.1 21.7 9.2 1.3 Morocco Rabat 99.1 99.1 99.1 98.7 98.1 79.3 82.4 87.0 100.0 99.9 99.7 99.7 99.6 86.3 89.1 93.9 95.9 98.8
Côte d'Ivoire 39.9 50.5 32.1 13.3 4.4 0.7 44.0 67.9 39.4 25.0 3.8 - Morocco Fes 99.6 74.4 99.5 99.4
Ethiopia 12.7 99.0 20.9 42.6 29.8 5.7 15.9 99.4 21.0 42.8 29.9 5.7 Morocco Marrakech 99.5 80.4 98.1 99.7
Gabon 68.1 56.1 39.2 13.2 3.5 0.2 82.3 66.2 46.3 15.6 4.1 0.3 Morocco Tangier 99.8 85.0 94.7 98.4
Ghana 33.5 80.4 46.7 27.9 5.6 0.2 36.4 69.6 43.5 21.2 4.8 0.1 Morocco Meknès 95.4 74.8 98.8 97.0
Guinea 23.4 79.6 34.9 34.1 9.4 1.2 27.9 72.3 31.7 31.0 8.5 1.1
Sub-Saharan Africa
Kenya 24.0 70.4 38.2 24.3 6.2 1.6 34.4 70.7 41.6 20.1 9.0
Angola Luanda 51.6 62.9 51.9 59.5
Madagascar 23.6 90.9 20.8 26.4 27.0 16.7 30.1 92.9 21.2 27.0 27.6 17.1
Benin Djougou 61.6 66.4 74.5 77.7 82.5 81.0 81.0 81.0 81.0 81.0 85.0 84.2 82.7 82.2 81.3 47.5 43.0 35.6 32.6 28.1
Malawi 11.6 94.6 57.1 33.0 4.5 15.1 91.1 55.0 31.8 4.4 -
Benin Porto-Novo 70.4 74.9 82.5 85.5 90.0 80.2 80.2 80.2 80.2 80.2 69.6 78.5 54.3 46.1 32.5 27.0 18.8
Mali 23.8 94.1 37.7 56.4 30.9 93.2 40.8 34.4 18.0 - Burkina Faso Ouagadougou 90.2 91.7 94.2 95.2 96.7 85.5 85.4 85.2 85.2 85.0 77.5 81.4 87.8 90.4 94.3 50.9 50.6 50.1 49.9 49.6
Mozambique 21.1 94.5 36.2 36.5 16.4 5.4 33.3 94.1 36.2 30.8 22.9 4.2 Cameroon Yaounde 92.1 92.2 92.6 92.7 92.9 84.8 86.8 90.2 91.5 93.5 83.3 83.7 84.5 84.8 85.2 11.8 37.8 81.2 98.6 98.6
Namibia 26.6 42.3 9.3 18.1 13.5 1.5 31.4 37.9 8.3 16.2 12.1 1.3 Côte d'Ivoire Abidjan 99.6 99.3 98.9 98.8 98.5 65.2 68.4 73.9 76.0 79.3 98.9 99.2 99.6 99.8 99.8 76.3 76.9 78.0 78.5 79.2
Niger 16.1 96.0 30.8 43.0 19.2 3.1 21.1 96.2 30.8 43.1 19.2 3.1 Dem. Rep.
of the Congo Kinshasa 86.9 46.5 85.5 78.2
Nigeria 35.0 80.0 31.4 35.1 11.3 2.3 44.9 79.2 36.8 30.7 9.9 1.7
Dem. Rep.
Rwanda 5.3 82.2 29.0 26.9 21.9 4.5 6.3 87.9 45.7 30.4 12.3 - of the Congo Butembo 21.6 55.7 70.1 82.5
Senegal 40.0 77.6 60.7 14.7 2.2 48.2 76.4 63.7 11.2 1.5 - Ethiopia Addis Ababa 66.7 64.5 98.4 48.1
South Africa 48.8 46.2 34.1 10.3 1.6 0.1 57.7 33.2 24.5 7.4 1.2 0.1 Ethiopia Nazret 22.7 61.2 85.7 34.9
Togo 28.5 80.9 55.7 21.4 3.3 0.6 33.9 80.6 55.5 21.3 3.3 0.5 Gambia Banjul 96.0
U. Rep of Tanzania 21.7 99.1 28.5 44.4 22.4 3.8 33.3 92.1 54.0 24.6 12.1 1.3 Ghana Accra 99.9 99.7 99.3 99.2 99.0 76.9 77.9 79.4 80.0 80.9 99.5 99.5 98.1 97.5 96.7 58.8 62.7 69.2 71.8 75.7
Zambia 39.4 72.0 32.1 22.8 14.0 3.2 39.8 74.0 46.5 20.8 6.4 0.2 Guinea Conakry 98.7 72.4 93.7 42.5
Lesotho Maseru 85.4 90.2 90.3 45.2
Zimbabwe 28.4 4.0 3.3 0.5 0.1 0.1 36.0 3.4 3.2 0.2 - -
Mali Bamako 100.0 98.0 83.1 77.1 68.1 77.1 77.1 77.1 77.1 77.1 87.4 88.0 88.9 89.2 89.8 38.9 41.3 45.2 46.7 49.1
Mozambique Maputo 83.2 81.7 96.7 46.3
Latin America and the Caribbean
Nigeria Lagos 88.2 90.9 95.3 97.1 99.8 60.9 60.9 61.1 61.1 61.2 94.9 94.9 94.9 92.3 88.5 100.0 95.8 85.3 81.1 74.8
Bolivia 55.6 70.0 45.0 24.9 62.9 61.3 26.3 20.4 11.3 3.4 Nigeria Ibadan 96.9 95.1 95.1 82.5 77.5 69.9 90.6 90.6 90.6 90.6 62.6 26.8 26.8 26.8 26.8 67.3
Brazil 74.8 45.0 32.0 11.0 1.9 0.2 81.7 36.6 26.0 8.9 1.5 0.1 Nigeria Ogbomosho 99.1 98.0 96.2 95.5 94.4 66.7 70.9 77.9 80.7 84.9 92.0 87.6 80.3 77.4 73.0 45.3 43.5 40.5 39.3 37.5
Colombia 68.7 26.0 19.9 4.4 1.3 0.4 75.5 21.8 18.2 3.0 0.5 0.1 Nigeria Zaria 99.1 99.0 98.9 98.8 98.7 46.7 51.7 60.0 63.3 68.3 96.2 97.1 98.5 99.1 99.1 52.9 54.1 56.0 56.8 57.9
Dominican Republic 58.4 56.4 42.4 12.2 1.7 0.1 66.0 37.6 28.2 8.2 1.1 0.1 Nigeria Akure 94.4 95.7 97.8 98.7 98.7 100.0 100.0 89.1 83.5 75.1 72.4 77.2 84.5 92.5 83.0 67.0 60.6 51.1
Guatemala 38.1 65.8 33.1 20.2 11.1 1.3 39.9 61.8 31.1 19.0 10.4 1.2 Rwanda Kigali 62.7 65.5 70.0 71.8 74.6 86.7 86.7 86.7 86.7 86.7 72.7 75.4 80.0 81.8 84.5 47.8 55.1 67.2 72.1 79.4
Haiti 29.5 84.9 67.0 15.5 2.4 36.3 85.7 67.6 15.7 2.4 - Senegal Dakar 97.8 98.7 98.7 98.7 98.7 69.2 70.0 71.4 71.9 72.8 97.1 96.0 94.3 93.6 92.6 59.0 63.9 72.2 75.6 80.5
South Africa Johannesburg 99.0 90.9 98.3 90.5
Nicaragua 53.1 80.7 23.8 23.7 21.1 12.1 56.5 80.9 33.7 29.1 17.9 0.3
South Africa Cape Town 97.2 85.7 98.8 94.7
Peru 68.9 60.4 20.9 19.3 14.1 6.1 73.1 68.1 32.6 20.3 11.9 3.3
South Africa Durban 85.1 72.3
South Africa Pretoria 99.0 90.9 98.3 90.5
Asia South Africa Port Elizabeth 57.8 79.9 61.4
Bangladesh 19.8 87.3 36.7 32.4 18.0 0.3 25.6 84.7 28.1 37.5 18.9 0.2 South Africa West Rand 81.9 98.4 83.4
India 25.5 60.8 37.8 20.1 2.9 27.9 55.5 39.5 15.2 0.8 - Uganda Kampala 69.6 74.1 81.7 84.7 89.3 62.7 62.7 62.7 62.7 62.7 75.6 87.3 87.3 55.0 56.6 59.2 60.2 61.8
Indonesia 30.6 32.2 18.8 11.2 2.1 42.1 23.1 18.0 4.2 0.9 - United Republic
of Tanzania Dar es Salaam 79.2 82.2 87.1 89.1 92.1 76.9 79.8 84.6 86.5 89.4 93.9 91.0 86.2 84.3 81.4 51.8 52.1 52.5 52.7 53.0
Nepal 9.0 96.9 19.8 35.5 34.6 7.1 12.2 92.4 40.2 37.4 14.8 -
United Republic
Pakistan 30.6 78.7 56.0 20.6 2.1 33.4 73.6 52.4 19.2 2.0 - of Tanzania Arusha 57.0 56.3 55.0 54.5 53.7 65.7 70.5 78.4 81.6 86.4 92.8 94.6 97.5 98.7 98.7 51.9 51.7 51.4 51.3 51.0
Philippines 48.8 54.9 16.7 18.3 14.4 5.5 59.4 44.1 32.9 8.9 2.0 0.2 Zambia Ndola 96.9 96.9 89.5 86.5 82.1 76.2 76.2 76.2 76.2 76.2 98.5 95.9 91.5 89.7 87.1 84.5 84.2 83.8 83.6 83.3
Turkey 61.2 23.3 18.7 3.7 0.8 0.1 66.2 17.9 15.5 2.0 0.3 0.1 Zambia Chingola 96.5 96.5 96.5 94.6 91.6 79.7 79.7 79.7 79.7 79.7 95.2 93.7 91.2 90.2 88.6 84.6 86.4 89.5 90.7 92.6
Viet Nam 20.3 60.5 37.9 15.7 5.6 1.3 24.5 47.4 29.7 12.3 4.4 1.0 Zimbabwe Harare 93.8 94.6 96.1 96.7 97.6 85.0 85.0 85.0 85.0 85.0 97.7 98.0 98.6 98.9 99.2 98.2 97.8 97.3 97.0 96.7
Yemen 22.8 67.5 34.3 18.6 12.2 2.4 25.0 65.1 33.1 18.0 11.7 2.3

194 195
TABLE 4 : PROPORTION OF URBAN HOUSEHOLDS WITH FINISHED MAIN FLOOR MATERIALS, SUFFICIENT LIVING AREA, SUSTAINABLE ACCESS TO
SAFE WATER AND IMPROVED SANITATION

Finished main floor materials Access to sufficient living area Access to safe water source Access to improved sanitation Finished main floor materials Access to sufficient living area Access to safe water source Access to improved sanitation

Country City 1990 1993 1998 2000 2003 1990 1993 1998 2000 2003 1990 1993 1998 2000 2003 1990 1993 1998 2000 2003 Country City 1990 1993 1998 2000 2003 1990 1993 1998 2000 2003 1990 1993 1998 2000 2003 1990 1993 1998 2000 2003

Latin America and the Caribbean India Amritsar 71.1 95.8 95.2 94.3 94.0 93.4
Brazil Sao Paolo 99.6 83.3 97.3 97.5 97.7 97.8 98.0 83.7 84.8 86.7 87.4 88.5 India Srinagar 77.1 96.4 97.0 98.1 98.5 99.2 68.9 74.5 83.9 87.6 93.2
Brazil Rio de Janeiro 99.6 89.7 95.6 96.0 96.6 96.9 97.3 67.5 70.2 74.6 76.3 79.0 India Jodhpur 77.9 99.2 99.0 98.6 98.5 98.2 77.4 82.5 91.0 94.4 99.5
Brazil Belo Horizonte 97.6 91.2 93.6 94.7 96.6 97.3 98.5 83.3 85.7 89.8 91.4 93.9 India Akola 65.8 88.0 89.9 93.1 94.4 96.3 53.5 60.1 71.0 75.4 82.0
Brazil Fortaleza 95.3 90.5 68.9 74.6 84.2 88.0 93.7 36.8 43.8 54.2 India Rajahmundry 80.2 83.6 88.4 96.4 99.6 99.6 45.7 49.3 55.2 57.6 61.2
Brazil Curitiba 96.8 96.1 94.9 95.9 97.6 98.2 99.2 55.4 61.6 71.8 75.9 82.0 India Yamunanagar 74.2 96.9 98.1 98.1 98.1 98.1 70.5 74.9 82.4 85.3 89.8
Brazil Brasilia 99.6 88.7 85.1 87.7 92.0 93.8 96.4 72.1 75.3 80.5 82.6 85.7 India Kharagpur 68.2 89.5 92.2 96.7 98.5 98.5 64.2 74.8 92.4 99.4 99.4
Brazil Goiânia 99.1 92.0 85.7 88.7 93.8 95.8 98.8 74.1 73.8 73.3 73.0 72.7 India Hisar 69.5 91.1 94.3 99.7 99.7 99.7 78.2 85.8 97.1
Brazil São José dos Campos 97.3 97.4 97.5 97.5 97.6 85.1 86.4 88.6 89.5 90.8 India Jalna 94.8 99.6 95.4
Brazil Nova Iguaçu 91.1 50.6 India Karnal 82.9 99.8 87.4
Brazil Ribeirão Preto 97.7 97.8 97.8 97.9 97.9 92.9 93.5 94.5 94.9 95.5 India Agartala 72.0 79.2 86.9 99.6 99.6 99.6 78.4 83.8 92.8 96.4 96.4
Brazil Vitoria 97.1 87.9 96.7 89.3 India Gadag-Betigeri 74.9 97.0 98.0 99.8 99.8 99.8 73.7 76.6 81.4 83.4 86.3
Brazil Guarujá 91.7 92.8 94.8 95.5 96.7 72.5 71.9 70.9 70.6 70.0 India Krishnanagar 73.3 94.0 95.1 97.0 97.8 98.9 70.7 74.6 81.1 83.7 87.6
Brazil Rondonópolis 78.6 81.1 85.3 86.9 89.4 16.6 20.1 25.8 28.1 31.6 Kazakhstan Shimkent 37.4 82.2 80.2
Chile Santiago 86.6 87.9 90.0 90.9 92.2 Kazakhstan Zhezkazgan 43.0 100.0 99.8
Chile Chillan 81.6 85.3 91.5 94.0 97.7 Pakistan Karachi 99.6 42.3 96.6 90.0
Colombia Bogotá 95.5 90.8 100.0 100.0 Pakistan Faisalabad 98.6 39.5 98.1 87.2
Colombia Medellín 99.9 93.8 100.0 99.7 Pakistan Islamabad 98.9 49.1 94.1 70.3
Colombia Neiva 96.9 91.2 100.0 99.5 Tajikistan Dushanbe 94.2 90.4 99.7 89.1
Colombia Valledupar 99.6 82.0 99.6 99.8 Uzbekistan Tashkent 99.7 97.1 100.0 90.7
Ecuador Guayaquil 95.9 96.2 96.6 96.8 97.1 51.6 51.0 50.1 49.7 49.1
Guatemala Guatemala City 80.0 71.6 82.6 83.8 85.7 86.5 87.7 32.7 44.7 64.6 72.6 84.6 South-eastern Asia
Mexico Mexico 91.2 92.5 94.7 95.6 97.0 80.7 83.2 87.4 89.1 91.6 Cambodia Phnom Penh 96.9 81.2 95.4
Mexico Guadalajara 96.7 97.2 98.1 98.5 99.0 96.4 97.1 98.2 98.7 99.3 Cambodia Siem Reab 96.9 57.9 45.8
Mexico Tijuana 67.8 74.8 86.5 91.2 98.2 57.0 63.2 73.5 77.6 83.8 Indonesia Jakarta 97.8 98.3 99.0 99.4 99.5 99.0 99.0 99.0 99.0 99.0 94.9 95.8 97.2 97.8 96.7
Mexico León 88.3 89.4 91.3 92.0 93.1 83.4 84.6 86.4 87.2 88.3 Indonesia Bandung 99.0 99.0 99.0 98.6 98.0 99.9 99.9 94.7 92.7 89.6 99.9 96.6 91.2 89.0 85.7
Mexico Culiacán 91.8 93.3 95.8 96.8 98.2 70.6 75.3 83.1 86.3 91.0 Indonesia Surabaja 98.8 98.8 96.8 95.4 93.2 97.8 97.8 92.9 90.9 87.9 89.0 89.0 89.0 81.2 69.6
Mexico Hermosillo 94.2 94.2 94.2 94.3 94.3 73.3 77.2 83.7 86.3 90.2 Indonesia Medan 99.8 99.8 99.8 99.8 99.8 95.3 96.2 97.8 98.4 99.4 98.8 98.0 95.4 92.4 88.0
Mexico Villahermosa 88.5 90.2 93.0 94.1 95.8 81.7 82.4 83.5 83.9 84.6 Indonesia Palembang 95.7 96.5 97.8 98.4 99.1 90.7 92.3 95.1 96.2 97.8 97.8 97.8 95.3 93.2 90.2
Uruguay Montevideo 90.1 91.6 94.2 95.2 96.7 94.5 Indonesia Ujung Pandang 94.5 95.9 98.1 99.0 99.8 99.4 99.4 99.4 99.4 99.4 76.9 83.2 93.6 97.8 99.3
Venezuela Caracas 85.4 87.3 90.5 91.8 93.7 88.0 95.2 96.4 96.9 97.6 Indonesia Bogor 95.4 95.9 96.7 97.0 97.5 93.9 94.0 94.3 94.5 94.7 62.8 71.2 85.2 90.8 99.1
Venezuela Maracaibo 82.0 84.6 88.8 90.5 93.1 88.0 88.0 88.1 88.1 88.1 Indonesia Surakarta 88.7 89.1 89.7 90.0 90.4 98.1 99.8 97.5 96.6 95.3 96.0 96.0 88.0 81.4 71.3
Venezuela Valencia 90.3 94.3 95.8 98.2 Indonesia Pekan Baru 99.9 99.9 99.9 99.8 99.7 97.2 97.2 97.1 96.2 94.9 99.5 99.5 95.1 92.2 87.7
Indonesia Denpasar 97.4 97.7 98.1 98.3 98.5 98.3 98.3 96.9 95.4 93.2 98.3 98.3 97.4 96.2 94.6
Eastern Asia 98.4 91.5 99.0 97.5 95.0 94.0 92.5 64.0 65.3 67.3 68.2 69.4 Indonesia Jambi 99.9 99.3 98.3 98.0 97.4 97.8 98.3 99.2 99.5 99.5 97.8 97.8 93.6 90.3 85.3
China Shanghai 92.3 Indonesia Purwokerto 75.7 77.2 79.6 80.5 82.0 88.9 88.9 86.1 82.5 77.0 53.9 58.7 66.8 70.1 75.0
China Beijing 92.3 Indonesia Kediri 79.4 82.9 88.8 91.2 94.7 92.8 94.6 97.7 98.9 99.0 49.0 55.9 67.3 71.9 78.7
China Guangzhou 92.2 Indonesia Palu 97.4 97.8 98.5 98.7 99.2 98.6 98.5 98.3 98.2 98.1 78.7 81.5 86.1 88.0 90.8
China Harbin 92.2 Indonesia Bitung 89.7 91.3 94.0 95.1 96.8 73.6 79.2 88.4 92.1 97.7 83.2 85.3 88.9 90.3 92.4
China Zhengzhou 92.3 Indonesia Jaya Pura 86.3 90.6 97.8 97.8 97.8 47.1 67.2 94.1 94.1 94.1 82.7 85.1 89.2 90.8 93.2
China Lanzhou 92.4 Indonesia Dumai 98.5 98.8 99.3 99.5 99.8 77.4 82.4 90.8 94.2 99.3 69.6 74.7 83.2 86.6 91.7
China Xuzhou 92.3 Myanmar Yangon 93.0 44.8 95.3 81.4
China Yulin 92.3 Philippines Metro Manila 78.4 78.8 79.4 79.6 79.9 65.8 67.8 71.0 72.3 74.3 83.8 86.8 91.9 93.9 96.9 85.6 88.9 94.2 96.4 99.6
China Yiyang 92.3 Philippines Cebu 51.1 54.1 59.0 61.0 64.0 63.1 66.5 72.2 74.5 77.9 66.5 73.8 86.0 90.9 98.3 79.0 81.0 84.5 85.9 87.9
China Yueyang 92.3 Philippines Cagayan de Oro 70.9 71.6 72.8 73.3 74.0 70.9 70.2 69.1 68.7 68.0 85.9 87.1 89.1 89.9 91.2 52.9 65.0 85.1 93.2 93.2
China Datong 92.3 Philippines Bacolod 41.8 46.3 53.9 56.9 61.4 69.6 70.1 70.8 71.1 71.6 84.2 84.2 84.2 91.3 91.3 56.0 62.3 72.8 77.0 83.2
China Leshan 92.3 Viet Nam Ho Chi Minh City 99.2 99.4 99.7 99.9 99.9 70.5 77.4 99.9 99.9 99.2 99.0 98.6 89.6 91.5 94.5 95.7 97.6
China Yongzhou 92.3 Viet Nam Ha Noi 95.1 96.3 98.4 99.2 99.2 75.7 82.2 96.2 97.1 98.7 99.4 99.4 58.4 67.8 83.3 89.6 98.9
China Chifeng 92.2 Viet Nam Hai Phong 98.1 98.0 97.9 97.8 97.8 84.3 95.9 98.0 98.5 99.0 99.2 99.0 1.0 1.0 69.3 81.7 100.0
China Huaibei 92.5 Viet Nam Da Nang 93.5 95.1 97.8 98.9 98.9 64.4 74.2 84.7 88.2 94.2 96.6 96.6 76.9 82.7 92.3 96.1 100.0
China Hegang 92.2
China Dandong 92.2 Western Asia
China Dezhou 92.4 Armenia Yerevan 98.9 99.4 93.6
China Anqing 92.3 Azerbaijan Baku 99.4 88.2 91.6 85.3
China Shaoguan 92.3 Iraq Baghdad 93.3 99.4 98.1
China Changzhi 92.4 Iraq Mosul 87.8 99.8 98.0
Mongolia Ulan Bator 98.4 44.1 97.0 75.3 Iraq Amara 88.8 93.1 88.8
Syrian Arab
South-central Asia Republic Damascus 99.7 99.1
Bangladesh Dhaka 71.0 60.2 99.5 90.4 Turkey Istanbul 99.8 99.8 99.7 99.7 99.7 92.8 93.6 95.1 95.6 96.5 98.7 95.5 90.0 87.8 84.5 99.2 99.2 99.2 99.2 99.2
Bangladesh Rajshahi 42.9 55.5 99.1 73.8 Turkey Ankara 99.8 99.8 99.7 99.7 99.7 91.3 93.6 97.4 98.9 98.9 99.8 99.8 97.4 96.4 95.0 98.9 99.1 99.5 99.6 99.8
India Mumbai 59.0 96.1 97.5 99.7 99.7 99.7 77.4 85.0 97.8 97.8 97.8 Turkey Izmir 99.2 99.4 99.7 99.8 99.8 96.1 95.8 95.2 95.0 94.6 98.6 99.1 99.1 99.1 99.1 99.0 99.2 99.3 99.4 99.5
India Kolkota 73.0 97.2 97.8 98.6 99.0 99.5 86.5 88.3 91.3 92.5 94.2 Turkey Bursa 97.1 98.2 98.3 98.4 98.4 96.3 96.3 96.3 96.3 96.3 92.0 92.0 92.0 88.8 84.0 98.1 98.1 98.1 97.4 96.3
India Delhi 73.3 99.6 99.5 99.2 99.1 99.0 81.1 85.8 93.7 96.8 99.0 Turkey Adana 99.8 99.8 99.8 99.8 99.8 80.5 83.6 88.7 90.8 93.9 97.7 98.6 98.6 98.6 98.6 99.9 98.8 97.1 96.3 95.3
India Hyderabad 78.9 97.4 98.4 98.4 98.4 98.4 77.3 80.3 85.1 87.1 90.0 Turkey Gaziantep 99.8 99.8 99.8 99.8 99.8 70.0 68.0 64.7 63.4 61.4 96.8 96.8 96.8 95.5 93.6 99.4 99.4 90.4 86.8 81.4
India Pune (Poona) 68.9 99.8 99.2 98.4 98.0 97.5 79.0 77.7 75.4 74.4 73.1 Turkey Kahramanmaras 78.8 78.6 78.3 78.1 78.0 82.6 87.1 93.8 88.6 92.9 92.9 92.9 92.9 95.4 85.7 69.6 69.6 69.6
India Kanpur 64.8 91.7 94.8 94.8 94.8 94.8 61.7 69.1 81.4 86.3 93.7 Turkey Antakya 99.5 99.5 99.5 99.3 99.0 89.5 91.5 94.8 96.1 98.1 92.7 92.7 92.7 89.8 85.4 99.4 99.4 83.9 77.6 68.3
India Jaipur 78.5 99.4 99.6 99.6 99.6 99.6 96.6 95.3 93.0 92.1 90.7 Turkey Aksaray 76.2 97.6 70.2
India Coimbatore 78.6 90.4 92.9 97.0 98.7 98.7 81.0 84.1 89.4 91.5 94.7 Yemen Sana'a 91.0 65.9 93.9 77.9
India Kochi (Cochin) 93.5 90.3 92.5 96.1 97.5 99.7 91.2 93.4 97.1 98.5 98.5 Yemen Aden 87.6 56.7 97.0 93.6
India Vijayawada 80.5 95.3 96.2 97.8 98.4 99.3 70.5 71.3 72.7 73.2 74.0 Yemen Taiz 91.5 58.0 85.6 77.1

196 197
TABLE 5 : PERCENT OF MALNOURISHED CHILDREN UNDER FIVE AND UNDER-FIVE MORTALITY RATES TABLE 6 : PERCENTAGE OF CHILDREN 12-23 MONTHS WHO RECEIVED MEASLES VACCINATIONS AND PERCENTAGE OF BIRTHS ATTENDED BY
SKILLED HEALTH PERSONEL
PERCENTAGE OF MALNOURISHED CHILDREN UNDER FIVE YEARS PERCENTAGE OF CHILDREN 12-23 MONTHS WHO RECEIVED MEASLES PERCENTAGE OF BIRTHS ATTENDED BY SKILLED HEALTH PERSONNEL
UNDER-FIVE MORTAITY RATES, DHS 1995-2003
(CHILDREN UNDERWEIGHT)

One Two Three+ One Two+ One Two Three One Two Three
Shelter Shelter Shelter All Shelter Shelter All Shelter Shelter Shelter All Shelter Shelter Shelter All
Urban Rural Non-Slum Deprivation Deprivations Deprivations Slum Urban Rural Non-Slum Deprivation Deprivations Slum Urban Rural Non-Slum Deprivation Deprivations Deprivations Slum Urban Rural Non-Slum Deprivation Deprivations Deprivations Slum

Africa Africa
Benin 17.8 25.3 11.6 18.9 24.6 30.0 24.3 134.0 175.0 98.0 114.0 171.0 142.0 Benin 75.3 64.1 87.4 72.7 68.0 55.1 65.1 79.9 59.0 94.9 78.0 64.6 44.7 62.1
Burkina Faso 20.5 40.3 22.4 35.2 38.3 43.1 39.4 136.0 202.0 129.0 128.0 197.0 151.0 Burkina Faso 73.1 53.3 69.7 66.9 52.6 49.6 54.2 87.7 30.5 86.3 57.0 30.6 22.4 32.4
Cameroon 14.3 25.0 12.8 14.9 23.5 27.8 23.8 111.0 160.0 87.0 118.0 141.0 129.0 Cameroon 67.6 49.3 74.0 74.7 53.3 39.5 50.5 82.1 44.8 85.2 84.0 50.8 34.6 49.8
Chad 31.6 40.6 26.4 29.4 36.7 40.4 39.0 Central African Rep 68.4 40.5 40.5 81.4 69.2 51.9 81.4 64.4 15.5 15.5 94.5 63.6 38.1 94.5
Comoros 25.0 26.1 20.2 20.0 30.2 46.0 26.8 Chad 38.9 18.6 60.1 55.4 28.4 18.3 22.2 12.4 1.6 30.5 15.7 4.8 2.4 3.5
Côte d'Ivoire 13.3 25.1 10.3 17.7 21.4 37.0 23.1 125.0 197.0 116.0 124.0 141.0 132.0 Comoros 63.0 63.5 85.7 67.9 57.0 37.5 60.0 64.9 35.9 72.8 53.3 27.9 22.1 38.0
Egypt 6.8 9.6 7.9 11.2 9.3 10.7 42.0 63.0 41.0 76.0 61.0 Côte d'Ivoire 82.0 58.8 83.3 79.8 61.9 45.9 62.8
Ethiopia 34.0 48.6 26.5 24.2 44.5 48.9 47.2 149.0 192.0 95.0 158.0 190.0 180.0 Egypt 96.0 95.3 96.4 92.1 95.5 52.1 96.4 86.7 59.0 76.6 43.1 60.7 29.0 76.6
Gabon 10.0 16.8 8.2 10.7 14.6 20.8 14.1 Ethiopia 63.1 22.3 88.6 71.0 37.5 21.6 26.4
Ghana 14.9 25.2 13.8 21.2 26.7 26.4 24.0 93.0 118.0 95.0 99.0 91.0 Gabon 61.1 37.1 65.8 58.1 47.0 31.8 48.7 89.6 63.1 93.0 84.9 74.7 59.5 76.2
Guinea 18.4 25.3 16.6 18.8 21.5 26.4 23.4 149.0 211.0 81.0 140.0 168.0 153.0 Ghana 85.8 81.8 88.3 83.1 82.5 73.8 81.7 78.8 29.5 79.7 46.6 26.8 25.3 36.1
Kenya 12.6 21.4 8.1 11.9 19.3 24.1 21.0 Guinea 66.9 46.7 80.1 65.8 60.1 41.0 51.5 16.8 6.7 32.9 16.0 11.5 5.4 8.8
Madagascar 35.6 41.0 12.5 34.0 40.8 44.1 40.2 Kenya 85.9 69.7 88.4 87.8 73.5 65.8 71.0 72.0 34.5 82.1 65.1 44.6 26.6 37.5
Malawi 12.8 27.3 8.7 12.6 26.0 30.6 25.8 Madagascar 60.8 42.0 95.9 65.7 41.8 50.6 45.5
Mali 20.6 37.2 15.5 20.5 35.9 36.9 34.2 185.0 253.0 131.0 155.0 229.0 197.0 Malawi 90.6 82.0 88.5 89.8 83.4 80.1 83.1 80.6 50.5 89.8 82.2 53.3 43.0 53.4
Morocco 6.5 13.9 7.4 12.1 15.0 15.3 13.9 38.0 69.0 37.0 41.0 64.0 46.0 Mali 70.8 41.3 77.7 68.4 48.3 39.5 47.2 72.2 7.9 80.3 56.7 20.9 9.0 19.8
Mozambique 20.0 28.2 9.5 14.4 20.9 31.6 26.4 143.0 192.0 79.0 107.0 177.0 146.0 Morocco 94.2 85.9 93.8 87.5 84.8 81.9 85.4 85.3 39.5 80.5 49.5 38.1 21.3 39.1
Namibia 16.5 25.9 13.5 21.0 27.8 26.7 26.5 Mozambique 93.0 47.1 95.1 92.3 76.7 43.0 57.0 80.6 30.1 96.7 85.6 59.5 25.5 40.2
Niger 35.3 52.4 13.0 30.9 49.6 52.6 49.8 Namibia 84.3 78.4 85.3 79.1 79.7 75.8 80.4 93.1 66.3 94.0 81.2 67.1 64.0 68.4
Nigeria 22.4 31.7 12.1 23.1 30.0 38.4 31.0 153.0 243.0 81.0 141.0 205.0 174.0 Niger 67.1 27.8 85.9 69.8 40.0 26.0 34.5
Rwanda 15.6 26.1 8.8 15.4 25.9 26.5 25.3 141.0 216.0 105.0 147.0 162.0 156.0 Nigeria 52.1 28.5 76.5 46.6 35.1 14.2 31.3 57.0 25.0 83.9 51.5 30.6 12.1 28.8
Senegal 89.0 165.0 52.0 96.0 124.0 102.0 Rwanda 89.9 86.3 90.7 91.4 86.1 86.2 86.7 65.7 20.0 82.7 57.0 24.9 18.4 23.9
South Africa 43.0 71.0 38.0 62.0 58.0 Senegal 80.9 29.6 78.0 66.3 35.1 21.4 43.3
Togo 16.1 27.9 16.9 21.8 26.7 29.4 25.5 South Africa 85.1 79.3 85.4 84.1 81.1 68.0 80.3 93.4 75.5 95.0 84.7 76.6 62.0 77.5
U. Rep of Tanzania 19.4 30.9 14.4 22.3 28.9 31.6 29.1 142.0 166.0 71.0 134.0 159.0 146.0 Togo 58.0 38.2 71.7 47.5 41.1 32.8 41.2 72.3 17.9 84.5 43.7 21.9 13.6 28.0
Uganda 12.4 23.6 7.4 16.0 23.2 28.1 22.8 101.0 163.0 51.0 87.0 136.0 107.0 U. Rep of Tanzania 90.3 75.3 96.3 88.7 78.5 73.0 77.9 76.8 26.4 85.0 77.5 31.6 25.2 35.0
Zambia 23.8 30.2 19.3 24.2 28.8 31.7 29.3 140.0 182.0 99.0 153.0 188.0 146.0 Uganda 68.4 55.3 62.3 61.1 55.8 54.3 56.7 35.5 15.9 43.4 25.4 17.3 10.9 17.5
Zimbabwe 7.5 15.6 8.5 11.8 14.4 19.5 8.5 69.0 100.0 69.0 66.0 72.0 69.0 Zambia 85.5 83.9 85.9 88.3 88.2 80.1 84.2 78.7 26.1 90.9 74.9 37.5 20.8 36.2
Zimbabwe 86.2 75.7 82.5 73.1 74.3 84.1 82.5 89.4 64.2 85.7 69.8 66.3 58.7 85.7
Latin America and the Caribbean
Bolivia 4.8 10.9 3.1 5.0 10.7 12.9 9.0 Latin America and the Caribbean
Brazil 4.6 9.2 3.2 5.6 7.3 19.1 7.5 49.0 79.0 34.0 55.0 90.0 64.0 Bolivia 66.5 60.2 67.5 64.6 59.2 64.3 62.4 77.4 36.3 88.8 69.0 43.7 25.5 48.8
Colombia 5.7 8.9 5.5 7.5 9.4 14.7 9.3 24.0 36.0 23.0 38.0 34.0 Brazil 90.2 76.5 91.7 89.2 78.5 73.8 83.9 92.0 72.7 95.3 88.9 78.4 55.1 81.6
Dominican Republic 3.9 7.2 3.8 9.7 8.2 14.8 9.6 Colombia 93.8 69.2 93.3 77.2 66.0 57.8 70.3
Guatemala 15.6 29.1 11.4 22.5 30.9 32.7 29.0 58.0 69.0 60.0 46.0 62.0 56.0 Dominican Republic 89.5 86.0 88.7 90.5 86.3 59.9 87.1 77.7 77.6 77.1 80.2 79.8 66.9 79.3
Haiti 11.5 19.1 8.9 15.4 17.2 25.8 19.1 Guatemala 80.8 80.4 94.5 70.1 79.1 72.9 74.7 66.1 24.7 80.4 43.0 21.9 14.3 25.3
Nicaragua 5.9 13.4 4.4 9.1 11.3 14.9 11.0 Haiti 60.9 50.2 70.7 54.8 51.4 37.7 48.9 49.3 9.2 50.6 25.3 10.2 5.5 13.7
Peru 3.2 11.8 1.5 4.8 9.1 12.4 9.4 Nicaragua 77.1 74.1 70.8 78.0 78.6 66.4 76.6 88.7 45.5 90.7 74.1 62.7 24.0 61.1
Peru 74.0 69.3 77.8 74.7 68.8 66.7 69.3 69.3 20.1 77.7 57.4 33.7 20.6 33.6
Asia
Armenia 2.4 2.8 2.1 4.9 2.1 3.8 37.0 59.0 38.0 37.0 32.0 36.0 Asia
Bangladesh 39.8 49.2 24.7 25.2 48.9 53.6 42.5 97.0 113.0 80.0 76.0 130.0 105.0 Armenia 76.0 72.1 79.9 69.2 79.9 55.0 99.1 94.5 99.5 95.0 96.3
India 38.2 49.3 33.0 49.8 52.8 50.2 65.0 111.0 48.0 88.0 80.0 87.0 Bangladesh 80.7 68.9 91.2 85.2 76.9 73.8 79.3
Indonesia 42.0 65.0 38.0 47.0 54.0 48.0 India 69.2 45.3 75.6 57.2 41.0 - 55.2 73.0 32.8 81.1 56.9 40.3 55.0
Kazakhstan 4.8 3.9 5.5 7.9 . 4.0 50.0 73.0 36.0 95.0 75.0 Indonesia 77.6 66.2 80.8 75.2 63.6 92.7 72.3 61.2 24.2 69.1 53.8 41.3 30.9 49.9
Kyrgyzstan 5.9 12.5 1.9 9.0 1.9 8.2 Kazakhstan 81.4 76.2 81.8 84.2 77.7 - 81.0 98.4 99.0 98.6 97.9 98.3 98.1
Nepal 33.1 49.5 27.6 26.8 42.2 43.7 37.0 Kyrgyzstan 83.7 84.5 87.4 81.1 87.4 81.9 99.3 97.8 100.0 98.7 98.8
Pakistan 32.3 44.4 29.3 40.7 46.4 41.2 94.0 132.0 86.0 113.0 106.0 112.0 Nepal 80.6 69.9 86.1 75.8 75.6 86.0 76.3 50.3 8.1 69.0 53.9 32.2 9.8 37.3
Philippines 30.0 52.0 26.0 36.0 47.0 39.0 Pakistan 64.6 43.6 67.8 55.0 59.0 N/A 55.4 40.3 6.9 47.6 21.8 14.7 21.0
Turkey 6.2 11.9 5.7 8.1 8.3 7.8 51.0 74.0 52.0 47.0 59.0 49.0 Philippines 81.8 77.5 83.2 75.5 82.0 - 77.2 79.0 40.8 83.2 67.3 51.9 54.5 64.2
Uzbekistan 16.6 19.7 9.8 18.5 25.5 23.5 19.5 52.0 57.0 39.0 56.0 68.0 58.0 Turkey 82.2 72.7 84.7 72.9 80.4 41.8 72.7 87.7 68.8 89.1 83.5 82.4 80.5 83.2
Vietnam 16.0 36.0 14.0 18.0 31.0 21.0 Uzbekistan 84.2 94.7 89.1 84.0 62.1 - 82.4 100.0 96.4 100.0 100.0 100.0
Yemen 24.5 30.7 21.3 29.8 31.4 29.5 117.0 141.0 93.0 124.0 161.0 141.0 Viet Nam 94.3 80.7 95.2 94.7 94.3 73.7 91.8 97.6 73.1 99.7 93.9 89.3 73.7 92.0
Yemen 68.6 37.1 72.6 59.6 63.0 - 63.5 45.8 10.0 55.4 38.5 31.0 20.7 35.1

198 199
TABLE 7 : PERCENTAGE OF CHILDREN UNDER-FIVE WITH DIARRHEA AND ACUTE RESPIRATORY INFECTIONS (ARI) TABLE 8 : EDUCATION; LITERACY RATES BY SHELTER DEPRIVATION

PERCENTAGE OF CHILDREN UNDER-FIVE WITH DIARRHEA, DHS 1995-2003 PERCENTAGE OF CHILDREN UNDER FIVE YEARS WITH ARI, DHS 1995-2003 WOMEN

One Two Three One Two Three One Two Three

Shelter Shelter Shelter All Shelter Shelter Shelter All Total Total All Shelter Shelter Shelter
Urban Rural Non-Slum Deprivation Deprivations Deprivations Slum Urban Rural Non-Slum Deprivation Deprivations Deprivations Slum Year Urban Rural Non-Slum Slum Deprivation Deprivations Deprivations

Africa Northern Africa

Benin 11.3 14.4 8.8 12.3 13.7 15.8 14.0 11.3 12.3 9.8 14.8 10.7 11.8 12.2 Egypt 2003 73.7 42.9 75.0 60.2
Burkina Faso 21.1 20.6 22.1 20.9 20.0 20.9 20.5 8.0 8.6 8.0 9.7 8.7 8.0 8.6 Egypt 2000 69.7 34.8 71.1 56.0
Egypt 1995 65.9 30.0 69.7 31.6
Cameroon 17.0 19.7 12.6 16.2 18.2 23.3 20.1 20.2 19.5 15.7 22.3 21.3 20.4
Morocco 2004 67.5 24.6 68.7 53.4
Central African Rep 19.5 24.8 24.8 13.6 21.8 24.2 28.1 28.2 28.2 24.8 30.8 28.9 24.8 Morocco 1992 57.6 12.0 60.7 35.9
Chad 22.2 21.3 16.7 21.0 21.8 21.5 21.6 12.6 12.7 10.9 12.1 11.4 13.2 12.7
Comoros 27.5 21.7 26.2 18.4 25.6 25.4 22.6 22.0 22.3 15.2 20.3 26.2 23.4 Sub-Saharan Africa
Côte d'Ivoire 16.9 23.6 17.9 23.4 22.0 20.0 22.0 14.6 17.2 14.7 14.3 15.6 23.3 16.6 Benin 2001 42.9 12.7 62.4 32.0 43.5 23.9 6.4
Egypt 16.8 20.2 18.1 21.3 21.2 23.7 11.1 9.7 10.4 10.2 9.1 16.3 10.4 Benin 1996 40.5 9.5 58.5 30.8 44.2 22.2 5.1
Burkina Faso 2003 52.7 5.8 57.2 42.1 45.0 38.8 13.8
Ethiopia 16.7 24.5 6.2 13.4 23.8 24.1 23.7 16.3 25.4 2.8 16.8 21.7 25.7 24.5
Burkina Faso 1999 47.0 3.6 82.7 45.2 49.6 32.5 33.3
Gabon 16.3 14.3 16.5 15.0 14.6 17.4 15.3 13.7 11.2 11.7 16.0 12.8 11.1 13.9 Burkina Faso 1992 43.7 5.9 80.0 40.9 47.7 34.2 16.3
Ghana 13.6 16.1 12.3 16.7 14.3 19.0 16.1 8.9 10.6 8.1 11.1 10.3 10.6 Cameroon 1998 80.4 53.9 88.6 72.9 80.6 69.7 44.3
Guinea 17.8 22.4 12.8 19.4 20.4 22.5 21.3 14.4 16.5 10.3 14.9 16.4 16.1 16.0 Cameroon 1991 69.6 41.4 85.6 64.2 71.7 52.6 34.3
Kenya 17.0 15.8 10.7 15.4 16.5 16.8 16.5 16.4 18.9 10.8 15.6 17.9 20.8 19.2 Côte d'Ivoire 1999 54.2 27.6 66.3 40.6 38.1 49.6 50.0
Madagascar 30.0 26.3 29.9 27.6 27.1 24.9 27.0 21.1 24.6 8.6 17.4 24.9 25.7 24.0 Côte d'Ivoire 1994 48.1 24.8 56.9 34.7 37.2 30.7 16.7
Ethiopia 2000 65.3 15.4 79.2 64.7 77.6 60.1 49.4
Malawi 14.3 18.1 12.0 13.3 17.7 19.4 17.7 15.7 28.3 17.7 19.7 26.7 29.9 26.9
Ghana 1999 72.7 47.2 76.6 68.5 70.8 52.5
Mali 13.1 20.2 12.9 12.0 17.7 21.7 18.9 9.3 10.1 12.2 8.9 9.0 10.5 9.8 Ghana 1993 62.3 30.8 68.8 58.1 61.8 46.4
Morocco 11.5 12.4 10.9 13.6 14.6 13.4 11.3 12.0 12.1 10.0 11.8 11.7 11.1 Guinea 1999 33.4 4.3 60.7 30.7 35.1 27.0 11.6
Mozambique 30.6 18.0 18.0 25.0 17.8 21.1 20.7 15.7 10.8 15.3 16.7 12.4 10.7 11.8 Mali 2001 35.8 5.7 47.7 30.4 35.2 21.4
Namibia 12.6 11.7 12.3 13.5 10.5 13.5 11.9 13.8 19.7 12.7 13.9 19.4 23.3 19.7 Mali 1996 31.2 4.1 69.4 29.6 41.5 27.4
Niger 31.6 39.0 28.8 28.3 38.9 38.7 37.8 14.2 14.2 22.0 10.9 13.9 14.7 14.1 Mozambique 2003 64.9 21.6 94.2 62.1 78.6 55.8 31.4
Mozambique 1997 62.8 23.3 89.9 59.2 73.5 53.1 29.5
Nigeria 14.5 20.7 9.6 14.1 20.3 23.5 19.9 7.8 11.4 6.5 9.2 11.3 11.1 10.7
Nigeria 2003 67.5 37.9 89.3 58.9 67.2 55.1 22.7
Rwanda 12.3 17.7 6.9 14.6 16.9 18.2 17.4 15.3 22.3 13.7 15.2 18.7 24.4 21.6 Nigeria 1999 70.4 45.6 87.8 62.6 72.0 55.2 29.1
Senegal 13.9 15.7 14.9 14.8 15.0 15.9 15.1 Nigeria 1990 66.2 27.2 88.9 58.4 64.5 49.5 38.5
South Africa 10.8 15.7 9.8 15.6 14.6 16.5 15.5 18.9 19.6 19.5 19.2 18.7 19.7 19.1 Rwanda 2000 86.0 62.0 90.8 83.4 87.7 80.9 77.1
Togo 27.0 32.3 17.7 30.0 32.1 33.6 31.7 18.8 20.7 19.3 17.1 20.7 24.6 20.3 Rwanda 1992 81.4 59.6 93.8 78.7 88.2 75.7 67.6
U. Rep of Tanzania 9.8 12.9 6.1 9.1 13.7 11.9 12.5 12.2 14.3 10.8 14.3 14.8 14.0 Senegal 1997 50.8 11.4 69.2 43.6 45.5 37.7 23.5
Senegal 1993 46.4 6.3 62.0 39.3 42.4 26.3 17.1
Uganda 15.5 20.1 10.6 19.5 20.4 19.0 19.8 18.6 23.0 14.1 24.2 22.0 22.3 22.7
South Africa 1998 96.2 87.9 97.1 93.0 93.3 91.5
Zambia 21.1 21.2 14.7 23.6 21.6 21.6 22.0 13.8 14.9 8.2 14.7 16.6 14.9 15.4 Uganda 2001 84.2 52.5 94.2 82.3 88.1 73.1 65.1
Zimbabwe 11.9 11.9 16.2 14.0 15.2 11.9 11.3 18.1 11.5 18.9 18.0 18.2 11.5 Uganda 1995 81.4 48.0 94.3 79.1 86.0 79.7 61.3
United Republic of Tanzania 1999 80.3 57.6 92.0 79.3 85.0 73.6 50.6
Latin America and the Caribbean United Republic of Tanzania 1996 82.6 60.2 94.0 81.4 88.9 77.1 60.9
Bolivia 21.1 24.1 19.8 22.9 23.0 24.8 23.4 22.8 21.4 23.4 23.9 20.5 20.3 21.7 United Republic of Tanzania 1992 78.5 56.7 93.8 77.1 86.9 72.1 61.7
Zambia 2002 78.8 48.4 87.0 72.1 73.7 69.2 70.3
Brazil 12.6 14.6 11.0 13.5 15.8 16.4 14.6 24.0 22.9 23.0 24.3 22.9 28.5 24.2
Zambia 1996 82.2 53.6 90.8 73.8 77.2 71.3 62.9
Colombia 13.2 15.6 12.7 17.1 23.2 16.7 Zimbabwe 1999 90.6 63.3 91.1 85.2 85.9
Dominican Republic 13.5 14.9 13.2 15.0 19.1 20.7 16.6 19.3 20.1 19.3 20.4 19.8 26.5 20.6 Zimbabwe 1994 83.2 53.4 85.2 72.2 78.3
Guatemala 12.8 13.6 12.6 11.1 13.5 16.4 13.6 17.6 20.1 16.4 19.0 20.3 21.2 20.2
Haiti 24.1 26.6 22.8 26.2 25.3 29.1 26.6 32.1 42.9 30.0 37.7 43.0 45.9 42.2 Latin America and the Caribbean
Nicaragua 11.7 14.4 9.8 13.3 14.2 14.2 13.9 28.5 33.2 27.0 27.1 34.5 35.1 31.8 Brazil 1996 94.2 80.8 96.2 91.6 92.8 88.2 82.4
Colombia 2000 88.0 53.2 89.4 73.2 75.6 63.5 43.8
Peru 13.6 17.6 10.8 16.0 18.6 16.8 17.4 19.8 20.6 18.0 20.2 21.8 20.9 21.1
Colombia 1995 97.4 88.6 98.0 89.2 93.4 82.7 61.9
Guatemala 1998 83.7 60.6 92.4 83.7 74.8 53.5 63.9
Asia
Armenia 7.8 7.8 7.8 10.6 7.8 8.4 11.5 11.4 11.8 7.3 15.8 11.8 9.5 Asia
Bangladesh 7.1 5.9 3.9 8.5 6.5 8.2 7.6 14.9 17.2 12.1 12.7 13.7 20.4 15.4 Bangladesh 1999 62.2 41.7 89.8 56.8 74.0 63.1 41.2
India 19.3 18.8 18.2 22.0 20.0 21.8 16.0 20.1 14.9 18.0 22.3 18.5 Bangladesh 1996 58.3 33.5 82.3 49.4 73.9 48.8 30.9
India 1999 62.2 27.6 73.0 45.6 45.5 46.0
Indonesia 11.2 10.8 10.5 11.9 13.8 12.3 14.9 17.2 6.7 9.1 8.1 8.7
Kazakhstan 1999 99.5 99.4 99.8 99.3 99.2
Kazakhstan 14.8 12.3 13.0 15.1 19.2 17.6 3.2 2.8 2.8 4.4 3.4 3.8 Pakistan 1990 50.6 11.0 58.8 28.2 30.0 14.9
Kyrgyzstan 15.1 18.3 14.8 16.6 14.8 15.3 2.5 5.0 2.5 2.6 2.5 2.4 Uzbekistan 1996 99.9 99.7 100.0 99.9 99.9
Nepal 16.6 20.7 15.4 15.7 17.3 25.4 17.4 23.8 22.7 24.4 24.5 22.5 25.5 23.5 Indonesia 2002 91.5 82.1 94.4 86.1 87.3 85.6 62.9
Pakistan 15.0 14.2 14.6 15.5 21.6 16.1 13.8 16.8 12.9 16.2 16.8 16.2 Indonesia 1997 92.8 77.9 92.8 93.6 88.9
Philippines 10.7 10.6 10.1 12.7 10.6 40.8 12.7 8.3 12.2 7.2 12.0 14.0 12.2 Indonesia 1994 89.8 72.5 89.8 91.9 82.2 72.2
Philippines 2003 97.8 93.8 98.5 94.6 94.8 93.5
Turkey 26.1 35.7 24.5 30.6 32.5 48.4 31.7
Philippines 1998 99.0 94.6 99.5 97.9 98.7 95.7 96.0
Uzbekistan 8.5 3.7 11.1 6.5 16.6 7.5 2.5 0.5 5.4 1.4 1.5 1.4 Viet Nam 2002 97.3 90.0 98.5 94.6 94.1 96.1 93.8
Viet Nam 3.5 13.0 3.7 1.4 9.2 2.9 14.0 20.7 14.4 11.4 22.5 13.1 Armenia 2000 99.8 99.4 99.8 99.8 99.8 99.6
Yemen 25.7 35.8 19.9 33.9 31.6 24.0 32.6 17.2 25.8 13.7 20.5 23.0 22.6 21.3 Turkey 1998 88.0 77.6 88.2 88.2 87.9 86.9 63.4
Turkey 1993 82.1 64.5 83.5 79.5 82.4 73.4 57.4
Yemen 1991 37.4 6.2 48.2 24.0 29.6 17.4 12.9

200 201
TABLE 9 : PERCENTAGE OF FEMALE AND MALE AGED 15-24 YEARS UNEMPLOYED BY SHELTER DEPRIVIATION

FEMALE MALE

One Two Three One Two Three

Shelter Shelter Shelter All Shelter Shelter Shelter All
Year Urban Rural Non-Slum Deprivation Deprivations Deprivations Slum Urban Rural Non-Slum Deprivation Deprivations Deprivations Slum

Africa
Benin 2001 14.4 5.5 20.1 13.4 7.9 9.5 11.3 47.5 28.7 59.3 45.0 28.9 21.8 38.7
Burkina Faso 2003 13.8 1.4 14.2 13.0 12.6 8.3 12.7 2.7 4.7 2.8 24.5 2.7
Cameroon 2004 31.0 18.5 28.3 33.6 37.2 14.1 33.1 8.3 10.4
Chad 1996 39.9 35.9 50.6 52.9 41.2 33.9 39.0 11.1 2.9 32.9 0.0 13.9 7.8 9.6
Comoros 1996 42.4 43.5 38.2 41.2 50.3 64.3 45.0 21.8 29.1 19.0 20.7 18.8 60.0 23.5
Côte d'Ivoire 1999 29.3 11.5 25.8 35.7 27.1 33.6 13.3 7.2 10.4 15.7 17.6 12.7 15.6
Gabon 2000 32.2 37.2 28.1 37.2 37.5 38.0 37.3 8.2 9.9 5.0 11.3 12.0 15.8 11.8
Ghana 2003 29.6 18.7 31.0 25.5 33.2 58.3 27.5 22.5 14.6 28.2 21.6 14.3 19.5
Guinea 1999 26.1 17.6 26.0 28.1 22.3 30.8 26.1 12.3 5.1 15.7 8.9 13.1 14.2 12.1
Kenya 2003 25.6 19.4 22.6 27.8 28.9 28.9 28.3 21.9 17.6 25.2 20.8 15.4 32.4 20.3
Madagascar 1997 25.4 17.3 22.3 27.3 27.7 21.5 25.6
Mali 2001 34.8 38.9 21.6 31.1 32.4 34.3 2.4 0.8 4.9 2.5 1.3
Morocco 2004 49.6 78.3 47.7 61.5 73.1 87.5 62.46
Mozambique 2003 46.1 19.1 20.4 49.2 49.7 40.9 49.0 22.0 24.5 13.3 21.1 22.7 27.4 23.3
Niger 1998 66.1 45.3 54.9 65.1 64.7 78.3 66.4 17.2 3.8
Nigeria 2003 29.2 38.0 23.7 27.9 35.2 40.3 31.8 15.7 14.6 23.1 12.3 14.9 19.3 13.8
Rwanda 2000 35.6 8.7 15.1 50.4 32.4 26.6 39.3
Senegal 1997 49.4 48.1 37.9 49.4 51.7 62.9 50.7
South Africa 1998 26.9 31.0 25.0 33.3 31.6 33.3 33.0
Togo 1998 14.3 16.1 13.5 15.5 11.7 14.5 14.8 11.0 9.8
Uganda 2001 39.3 21.4 25.2 41.0 39.9 41.3 41.8 11.6 7.6 7.9 20.3 9.6 5.3 11.7
United Republic
of Tanzania 1999 32.9 14.9 17.0 31.5 40.6 25.0 34.7 10.5 7.1 13.7 9.7 13.1 1.6 10.3
Zambia 2002 46.1 38.2 46.0 49.5 41.6 48.1 47.3 19.3 17.5 19.9 14.3 24.0 16.3
Zimbabwe 1999 40.1 36.5 38.8 36.0 52.8 43.4 30.7 21.4 34.1 4.0 6.1

Latin America and the Carribean

Bolivia 2004 11.2 30.2 6.4 10.6 18.9 29.4 13.5 3.5 4.6 3.7 3.7 3.2 3.1
Brazil 1996 18.7 35.6 17.5 18.8 22.3 41.2 20.1 2.8 1.0 3.5 2.9 0.5 2.1
Colombia 2000 21.9 42.6 21.0 30.8 37.7 25.0 32.0
Guatemala 1998 40.3 57.8 30.5 49.6 69.5 71.4 57.3
Haiti 2000 29.1 33.4 23.3 32.8 40.6 39.2 34.7 14.9 12.5
Nicaragua 2001 26.6 60.5 17.2 26.2 35.5 57.2 31.3
Paraguay 1990 13.1 43.4 8.5 18.0 15.6 24.7 17.7
Peru 2000 17.6 30.6 14.2 19.4 22.5 31.2 22.4

Asia
Kazakhstan 1999 27.7 38.9 19.6 31.6 30.3 39.2 32.5 19.7 37.4 15.5 12.1 31.0 43.7 22.7
Krygystan 1997 26.1 47.2 16.2 14.0 32.8 54.8 29.4
Nepal 2002 45.2 22.7 61.0 45.3 34.7 0.0 35.1
Philippines 1998 15.3 30.6 12.6 19.8 28.6 33.3 22.2
Philippines 2003 17.6 31.5 23.5 14.3 23.3 34.0 25.9
Turkey 1998 49.5 41.8 51.3 47.3 32.4 33.3 45.5 1.2 15.5 2.2
Uzbekistan 1996 42.7 46.8 29.5 40.4 47.2 45.7 45.2
Viet Nam 2002 16.7 10.4 18.8 12.5 16.7 12.5

202